Identification, characterization, and molecular cloning of the immunodominant Babesia bovis merozoite surface protein Bv42

MISSING IMAGE

Material Information

Title:
Identification, characterization, and molecular cloning of the immunodominant Babesia bovis merozoite surface protein Bv42
Physical Description:
vii, 86 leaves : ill. ; 29 cm.
Language:
English
Creator:
Hines, Stephen Alan, 1955-
Publication Date:

Subjects

Subjects / Keywords:
Cloning, Molecular   ( mesh )
Babesia -- genetics   ( mesh )
Membrane Proteins   ( mesh )
Pathology thesis Ph.D   ( mesh )
Dissertations, Academic -- Pathology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1989.
Bibliography:
Bibliography: leaves 74-84.
Statement of Responsibility:
by Stephen A. Hines.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001140444
oclc - 22253070
notis - AFN9788
System ID:
AA00009079:00001

Full Text












IDENTIFICATION, CHARACTERIZATION, AND MOLECULAR CLONING OF
THE IMMUNODOMINANT BABESIA BOVIS MEROZOITE SURFACE PROTEIN
Bv42













By


STEPHEN A. HINES


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


1989














ACKNOWLEDGEMENTS


I want to thank some of the many people who contributed

to this work and helped to make my time at the University of

Florida memorable. First, it was a unique opportunity to

have had both Guy Palmer and Terry McElwain as major

professors. I hope that Guy and Terry know my gratitude,

respect, and true affection. The only problem in working

with people of their caliber is that you're left with a high

standard to measure up to. Although the other members of my

advisory committee (Tony Barbet, Lindsey Hutt-Fletcher, and

Ward Wakeland) were not as closely involved day-to-day with

my research, each of them has significantly influenced my

development as a scientist and I thank them sincerely.

I also thank my friends Drs. Richard and Catherine

Crandall who supported me when I first entered graduate

school and encouraged an interest in parasite immunology.

Likewise, Dr. Mike Burridge, Chairman of the Department of

Infectious Diseases in the College of Veterinary Medicine,

has done much to ensure the success of my work and the

project I was involved with.














The monoclonal antibodies and monospecific rabbit

anti-serum against Bv42 were produced and originally

characterized at Washington State University. My thanks

goes to Travis McGuire, Lance Perryman, William Davis, Will

Goff, and David Reducker who in collaboration with my major

professors developed and provided these valuable reagents.

During my tenure as a graduate student, I have worked

alongside some remarkable people. In addition to Terry and

Guy, these were the people who made work fun and helped me

look forward to each new day. Pam Kaylor, Dr. Jennifer

Garber, and Raina DeVenere were technicians who facilitated

my work and became good friends. Stephen Kania and Vishnu

Mishra shared lab space, an interest in infectious diseases,

mutual concern for each other, and some memorable laughs.

Kelly Sloan, Lili Sotomayor, Jane Farrell, and Teri Harty

were our veterinary student "elves" who helped so much with

the animal and Babesia culture work. These four vet

students have been treasures; they reminded us of why we

were interested in veterinary medicine and brought sunshine

into the lab every summer.

Lastly, I want to thank my wife Melissa Trogdon Hines,

D.V.M., Ph.D., without whom nothing else would matter very

much.


iii













TABLE OF CONTENTS
page
ACKNOWLEDGEMENTS ......................................... ii

KEY TO ABBREVIATIONS................................... V

ABSTRACT.......... ............ .............. ......... vi

CHAPTERS

1. LITERATURE REVIEW AND RESEARCH STRATEGY

Literature Review
Introduction and Significance ............ 1
Background -- Life Cycle and Disease..... 3
Current Methods of Control ............... 7
Experimental Nonviable and Semidefined
Immunogens........................... 12
Recombinant Babesia bovis Antigens....... 19
Summary ................................ 21
Research Strategy.............................. 23

2. MOLECULAR CHARACTERIZATION OF Babesia bovis
MEROZOITE SURFACE PROTEINS BEARING EPITOPES
IMMUNODOMINANT IN PROTECTED CATTLE

Summary......................... ....... ...... 27
Introduction.................................. 28
Materials and Methods......................... 29
Results....................................... 34
Discussion................................... 38

3. MOLECULAR CLONING OF THE GENE ENCODING THE
IMMUNODOMINANT Babesia bovis MEROZOITE SURFACE
PROTEIN Bv42 AND ITS EXPRESSION IN Eschericia coli

Summary ..................................... 50
Introduction................................... 51
Materials and Methods ........................ 52
Results....................... ... ........... 57
Discussion.................................... 60

4. SUMMARY / CONCLUSIONS........................... 70

REFERENCES..................................... .......... 74

BIOGRAPHICAL SKETCH ..................................... 85













KEY TO ABBREVIATIONS


PCV, packed cell volume

IFA, (indirect) immunofluorescence assay

MASP, microaerophilous stationary phase

(Babesia bovis cultures)

IV, intravenous

iRBC, infected erythrocytes

PBS, phosphate buffered saline

NP-40, Nonidet P-40

PMSF, phenylmethyl-sulfonylfluoride

nRBC, normal (uninfected) erythrocytes

VBS, veronal buffered saline

SDS-PAGE, sodium dodecyl sulfate polyacrylamide

gel electrophoresis

kDa, kilodalton

kb, kilobase











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

IDENTIFICATION, CHARACTERIZATION, AND MOLECULAR CLONING OF
THE IMMUNODOMINANT BABESIA BOVIS MEROZOITE SURFACE PROTEIN
Bv42

By

Stephen A. Hines

December 1989

Chairman: Terry F. McElwain
Major Department: Department of Pathology and Laboratory
Medicine, College of Medicine


Eight surface-radioiodinated merozoite proteins from a

cloned, pathogenic isolate of Babesia bovis can be

immunoprecipitated by antibody from cattle that are

completely protected against clinical babesiosis. Among

these eight surface proteins, the 55 and 42 kDa molecules

are biosynthetically labeled with 3H-glucosamine. The 42

kDa glycoprotein can also be labeled with 3H-myristic acid

and partitions exclusively into the detergent phase in

Triton X-114 extracts, indicating that it is an integral

membrane protein and suggesting that it is anchored by a

glycosylphosphatidylinositol moiety. Antibody mediated

protection against B. bovis merozoites most likely requires

a high level of circulating antibody to ensure antibody-

merozoite binding during the parasite's brief

extraerythrocytic phase. Antibodies in diluted sera











selectively recognize the 120, 85, 55, and 42 kDa surface

proteins. Only the 42 kDa integral membrane protein is

reactive with serum antibodies diluted > 1:16,000. Thus, we

hypothesize these immunodominant proteins, especially the

transmembrane 42 kDa glycoprotein, are important to the

induction of the protective immune response and are

candidates for an improved vaccine against babesiosis.

A 0.9 kb cDNA fragment encoding the 42 kDa

immunodominant merozoite surface antigen Bv42 was cloned

into the unique Eco RI site of the bacteriophage vector

Lambda ZAP II. The 45 kDa recombinant protein expressed in

Escherichia coli accounted for greater than 95% of the

native 42 kDa glycoprotein, including the epitope(s) defined

by monoclonal antibodies that react with the surface of live

merozoites. Antibodies prepared against recombinant Bv42

immunoprecipitated the native 42 kDa glycoprotein from

surface labeled and metabolically labeled B. bovis.

Further, antibodies against the recombinant protein bound to

and removed the 42 kDa protein recognized by surface

reactive monoclonal antibody Babb 35A4 from a preparation of

native, 35S-methionine labeled parasite antigen. The

cloning of the gene for Bv42 will now allow us to test the

ability of an immune response against this specific

merozoite surface antigen to protect cattle against

babesiosis.


vii














CHAPTER 1

LITERATURE REVIEW AND RESEARCH STRATEGY



Literature Review

Introduction and Significance

The arthropod-borne, hemoparasitic diseases of domestic

animals are significant obstacles to improved meat, milk,

and fiber production in the lesser developed tropical

nations, and represent a primary barrier to the importation

of more productive breeds of food animals [1]. Babesiosis

is a protozoal disease that is of critical importance to

cattle living in or bound for the tropical and subtropical

regions of the world. The two most important species,

Babesia bovis and B. bigemina, are estimated to endanger 500

million cattle between the thirty-second parallel north and

fortieth parallel south of the equator, where their

geographic distributions correspond to that of their vectors

[2,3].

Economic loss in babesiosis is attributable to high

mortality, which is in excess of fifty percent for

susceptible Bos taurus cattle, and to diminished production

of milk and meat in animals recovering from acute infection

and abortion. Additional losses include the high cost of













quarantine and other measures to control the spread of the

disease, as well as the costs associated with loss of market

and inability to import high yielding Bos taurus breeds [2].

In Australia, tick-borne disease costs the government and

cattle producers an estimated 42 million dollars

(Australian) per year [2]. Economic losses are also

significant in many Latin American countries, but it is

difficult to determine the economic cost of babesiosis in

much of the developing world due to lack of epidemiological

data [3].

In the U.S., eradication of Boophilus ticks and

babesiosis is estimated to save the country 500 million

dollars per year, but the potential for re-introduction of

Babesia-infected ticks from Mexico, the Caribbean, and the

Southern Hemisphere via illegal livestock importation

remains an ominous possibility [2]. The potential for

catastrophic outbreak even in areas where tick-borne disease

appears to be controlled was illustrated by the epizootics

that occurred during the Zimbabwean pre-independence war in

the mid-1970's. Nearly one million cattle died of

hemoparasitic disorders following the disruption of short-

interval dipping used to control cattle ticks [4].

In Zimbabwe and many of the endemic areas, B. bovis, B.

biqemina and the rickettsia Anaplasma marQinale can share











3

the same tick vector and have vectors with shared geographic

ranges. Infections often occur in combination, resulting in

synergistic pathogenicity and a disease complex collectively

called "tick fever." Control of this complex, including

attempts to control by vaccination, will ultimately require

a strategy directed against all three organisms.

Our long-range goal is an antigenically-defined subunit

vaccine against Babesia bovis. The protection afforded by

recovery from primary infection indicates that a vaccine is

an achievable goal. Earlier work with killed, semi-defined

antigen preparations illustrates the potential for a subunit

vaccine that could induce a protective immune response in

susceptible cattle.



Background--Life Cycle and Disease

Transmission of bovine babesiosis is by a variety of

ticks, of which Boophilus species are most notable.

Susceptible ticks become infected when they acquire a blood

meal from an infected cow and can transmit the disease to

other cattle only after the completion of a developmental

cycle that begins in the tick midgut [5]. Here the majority

of parasites in the imbibed blood die while a small

percentage of unique intracellular forms, the putative













gametocytes, begin a metamorphosis. Although fusion or

fertilization of gametes has not yet been observed, the

morphologic changes that occur suggest a sexual process

[6,7]. The result is a spherical body that is transformed

into an elongate ookinete. Invasion of gut epithelium is

followed by asexual division and the production of

sporokinetes (sporogony). These motile kinetes are released

by gut epithelium and disseminated via hemolymph to enter

salivary gland cells, muscle fibers, haemocytes, ovarian

cells and oocytes. Further asexual division in these cells

produces more kinetes which remain infective only for the

tick cells. Dormancy of the parasite in tick oocytes allows

for transovarial transmission to subsequent generations.

When progeny ticks mature and begin to feed, sporogony again

occurs in a manner similar to that following primary

infection in the adult. It is the invasion of salivary

gland cells by sporokinetes that initiates the form of

asexual reproduction that produces sporozoites which are

infective for cattle. Differentiation to the sporozoite

requires a feeding or temperature stimulus and the parasite

is then transmitted via saliva during tick feeding.

Following inoculation into cattle, sporozoites of

Babesia directly bind and enter erythrocytes. The













sporozoite becomes an intracellular feeding stage and

subsequently undergoes binary fission (merogony) to form two

to four daughter cells (merozoites). Merozoites exit the

infected erythrocyte causing host cell destruction and are

immediately infective for other erythrocytes, thereby

perpetuating the cycle of red blood cell invasion and

destruction.

The replicating population of Babesia in a host is

heterogenous due to the presence of subpopulations that

differ in antigenicity, transmissibility and virulence.

Likewise, there is evidence for antigenic differences

between isolates and for selection of subpopulations by

passage in both invertebrate and mammalian hosts [8,9].

Clinically, bovine babesiosis tends to be acute and

life-threatening, especially in Bos taurus cattle which are

more susceptible to infection than Bos indicus [10,11].

Hemolysis due to intracellular parasite multiplication and

subsequent erythrocyte destruction peaks at 7 to 20 days

post-infection. Accompanying signs include anemia, fever,

icterus, roughened hair coat, dehydration, and depression.

Pregnant animals may abort and hemoglobinuria is common.

B. bovis-infected erythrocytes sequester in visceral

vessels, notably cerebral and renal capillaries. Peripheral













blood parasitemias may remain below 1 percent, while

parasite rates in the cerebral capillaries typically exceed

90 percent [10]. Neurologic signs are attributed to the

blockage of cerebral capillaries by parasitized erythrocytes

and to anemic hypoxia. Also with B. bovis infections,

significant vascular changes occur when total parasite

numbers are still low and large scale hemolysis has not yet

developed [11]. Parasite-produced proteolytic enzymes are

released into the blood stream during the exit of merozoites

from infected red cells and activate certain plasma

proteins, including kallikrein and components of the

complement and clotting cascades [10]. The result is

activation of potent vasoactive mediators and a profound

hypotensive shock.

Cattle that survive the acute phase of babesiosis

typically become carriers whose blood is directly infective

to non-immune animals and tick vectors [11]. Although the

parasite is maintained in the cattle and tick populations

via these asymptomatic carriers, persistently infected

animals are also resistant to clinical disease if they are

challenged a state known as premunity or "co-infectious

immunity" [10]. Prolonged sterile immunity can be seen in

cattle that eliminate the organism either spontaneously or

due to antibiotic treatment [12,13]. This indicates that











7

living parasites need not be present in the blood stream for

animals to be resistant to disease [10]. The protection

engendered by infection is strong evidence that an effective

vaccine is an achievable goal.


Current Methods of Control

The United States' eradication of Babesia via its tick

vector has proven to be an isolated experience not likely to

be repeated on a large scale [2]. Current strategies to

control babesiosis in endemic areas are a combination of

vector reduction and immunoprophylaxis.

Vector Reduction Tick control by interval dipping and

exclusion of vectors from areas of animal congregation

remains an important mainstay in the prevention of

hemoparasitic diseases [3]. In many developing tropical

countries, acaricide programs consume a sizable percentage

of the national veterinary services budget [14]. These

programs also have important health considerations for

people involved in dipping, but they can effectively

decrease losses due to arthropod-borne diseases. The

paradox is that reduction of the tick population in endemic

areas not only retards the rate of Babesia transmission, it

also increases the number of susceptible cattle [4]. This

occurs in part because calves have a relatively high













resistance to clinical babesiosis for several months after

birth. Resistance is associated with poorly defined host

factors and with passive transfer of colostral antibodies

from immune dams [15]. Infection during this period

typically results in mild or inapparent disease and the

prolonged protection afforded by subclinical infection

(premunition) [15]. Effective methods of tick control

decrease the frequency of calfhood infection and therefore

also reduce the number of cattle protected by natural

infection. Subsequent changes in tick control practices in

endemic regions can upset the balance between numbers of

susceptible cattle and infected ticks, and result in

disastrous epizootics [4].

Immunoprophylaxis Vaccination is the logical method to

protect individual cattle and prevent outbreaks, especially

in highly susceptible animals that originate in non-endemic

areas or areas where Babesia infection rates are limited by

effective vector reduction programs. An effective

immunization program would increase the number of resistant

animals in an endemic area, thereby decreasing the required

dipping frequency and its associated economic burden while

restoring the enzootic stability needed to withstand

possible breakdowns in the control system [15].













Currently practiced methods of immunoprophylaxis use

premunization whereby resistance to the disease is

associated with the presence (at least transiently) of the

parasite in the bloodstream. Premunition is achieved by (I)

deliberate infection with an attenuated (non-pathogenic)

strain, or (II) drug cure of an infection after a

significant parasitemia has developed.

In Australia, the regional government in Queensland

annually distributes nearly one million doses of an

attenuated, live B. bovis "vaccine" [16]. The vaccine is

produced by multiple (20-30) syringe passages through

splenectomized calves and is administered as a subcutaneous

dose of infected erythrocytes [17,18]. The ensuing immune

response in vaccinates does not prevent multiplication of

challenge organisms (108) in erythrocytes or some early

deviations in packed cell volume or rectal temperature after

experimental challenge. However, these changes are reversed

by about the sixth day post-infection and most animals

recover without developing overt clinical signs [19].

Protection is similar to that conferred by recovery from

infection with a virulent strain transmitted by infected

ticks or direct inoculation of infected blood. Like natural

infection, protection persists for at least four years and

is effective against virulent heterologous parasites [19].













The Australia vaccine also has a number of serious

disadvantages [20]. These include the following:

1) infected blood has a shelf life of only six days;

2) variable attenuation results in clinical disease in a

minority of recipient cattle; 3) the vaccine is a blood

product raised in calves and can serve to transmit a number

of infectious agents including other hemoparasites and

Bovine Leukosis Virus; 4) the large mass of host components

in the inoculum can sensitize the vaccinates to bovine

erythrocytic proteins and predispose to neonatal

isoerythrolysis in progeny calves; 5) passage through non-

splenectomized cattle (such as the vaccinates) can result in

a reversion to virulence thereby maintaining virulent

parasites in the population for transmission to susceptible

individuals; 6) production and distribution requires a large

centralized facility and a cold-chain; and 7) the product is

not suitable for Babesia-free regions, where its use could

result in introduction of the disease.

More recently, investigators have reported that cattle

were protected against babesiosis after infection with an

attenuated strain of B. bovis produced in vitro [21,22]. In

previously unexposed cattle, the tissue culture-adapted

strains produced mild clinical signs and moderate changes in

erythrocyte packed cell volume. Production of a vaccine













strain of Babesia in culture would improve quality control

and decrease the need for splenectomized calves but would

not overcome many of the limitations inherent to

premunization using infected blood. Furthermore,

cultivation of hemoparasites on a large scale would be

difficult.

In Israel and elsewhere, the shelf life of an

attenuated inoculum has been extended by storage in liquid

nitrogen but the method is unwieldy [23]. Other problems

with a blood-derived, living product remain, and the low

survival rate of B. bovis after freeze/thawing necessitates

that large number of parasites be present in each dose -

thereby increasing both the mass of bovine protein

inoculated and the cost per vaccinate.

The second method of premunization involves infection

of cattle with attenuated or virulent strains of Babesia

followed by drug cure, often using the chemoprophylactic

imidocarb [24-26]. This technique is labor intensive and

requires skilled professionals to treat cattle only after a

protective immune response has had time to form but before

acute, life-threatening disease has occurred. A sterilizing

treatment administered early in the course of the infection

produces a less protective immune response than if the

treatment is given after a high parasitemia [12,24]. In













addition, chemotherapeutic drugs against Babesia are both

expensive and toxic.

Clearly, current methods to immunize cattle against

babesiosis are seriously flawed, but the solid protection

produced by these techniques is encouraging.



Experimental Nonviable and Semidefined Immunogens

Recent work on bovine babesiosis has focused on using

killed or semi-defined antigens to induce protective

immunity. These antigens can be classified as 1) killed

whole organism vaccines derived from infected blood,

2) antigens purified from parasitized erythrocytes by

affinity chromatography, and 3) soluble antigens from

continuous MASP cultures of B. bovis. Immunization with

such preparations does not prevent infection but can

significantly reduce morbidity and mortality. None of these

products is likely to become a satisfactory vaccine, but the

work clearly demonstrates the potential for a subunit

approach.


Whole organism vaccines

A freeze-dried suspension of killed Babesia bovis has

been shown to partially protect cattle from intravenous

challenge with 106 or 107 infected erythrocytes [27]. The













suspension was produced by lysis of infected cells in

distilled water and recovery of parasites by centrifugation.

Protection was manifested by fewer clinical signs and

decreased erythrocyte destruction. Subsequently, a crude

babesial antigen called infected erythrocyte antigen (IEA)

was prepared by disruption of concentrated, parasitized

erythrocytes in a French pressure cell [28]. Subcutaneous

inoculation of this product in Freund's complete adjuvant

was shown to protect Bos taurus cattle (n=4) against a

heterologous strain as effectively as recovery from a tick-

transmitted infection 12 months previously. Protection was

less effective when the challenge dose was increased from

106 to 108 infected erythrocytes. These studies indicated

that at least partial protection could be conferred with

killed antigen and suggested that antigenic variants were

not a formidable barrier to immunization. However, the

experimental preparations tested were also contaminated with

host erythrocyte antigens leading to the development of

isoantibodies [28].

Considerable effort has been directed toward

fractionation of crude babesial antigen and demonstration of

a protective immune response derived from a semidefined or

purified component [29-36]. At least two subfractions of

infected erythrocytes have been shown to induce a protective













immune response that is as effective as that produced by

crude antigenic material [30]. The persistent problem of

eliminating host protein from vaccine preparations has been

complicated by the discovery that some babesial antigens

complex with host protein.




Affinity purified immunogens

Three B. bovis antigens with molecular weights of 1300,

180, and 44 kDa were purified from infected blood using

monoclonal antibody affinity chromatography [37]. These

antigens could be specifically localized to the parasite or

infected erythrocyte by immunofluorescent staining. When

splenectomized calves were immunized twice with the

respective antigens in Freund's adjuvant, only the 44 kDa

antigen conferred protective immunity against a homologous

intravenous challenge of 103 organisms. Subsequently the

same 44 kDa-reactive monoclonal antibody was used to isolate

a 29 kDa antigen that had been further purified by gradient

gel electrophoresis [38]. Two subcutaneous injections of

this antigen in FCA induced a protective response in nine

nonsplenectomized adult cattle that were challenged with 10'

infected erythrocytes of the homologous, virulent strain.

Whereas none of the vaccinated animals showed clinical











15

signs, three of the five unvaccinated controls were severely

affected. No animals in either group died but the

vaccinated cattle had significantly less change in PCV, body

temperature, and parasitemia.

Based on the relative protection of splenectomized

calves, the authors judged the affinity-purified 29 kDa

antigen to be as protective as crude antigen. The results

in intact adult cattle were much harder to interpret

considering variations in challenge dose, quantity of

immunogen and strains of B. bovis used for antigen

extraction and challenge. The monoclonal antibody used to

isolate the 29 kDa antigen is a labile IgM that cannot be

separated from contaminating ascites proteins without

severely affecting its biological properties. The resulting

presence of haptoglobin in the immunosorbent allows for

binding of bovine hemoglobin and subsequent elution with the

babesial antigen [38]. Native 29 kDa antigen is also

apparently completed to large host molecules that may act as

carriers and even alter immunogenicity. Production of

sufficient purified antigen clearly remains a significant

problem in testing experimental vaccines. Nevertheless,

this work demonstrates the potential of a defined antigen

approach to develop effective immunoprophylaxis in bovine

babesiosis.













MASP culture derived antigens

Candidate protective antigens of Babesia bovis have

also been studied as a by-product of recent advances in in

vitro cultivation by the microaerophilous, stationary phase

(MASP) technique [39,40]. Immunoelectrophoresis of culture

supernatants demonstrated at least three soluble antigens

with molecular weights in the range of 37-40 kDa [41,42].

These may be identical to three antigens that were

previously identified in the lytic extracts of infected

erythrocytes [41,29]. When monospecific rabbit antibodies

against the purified culture-derived antigens were used in

an immunofluorescence assay (IFA), two of the antigens (1

and 2) could be localized to the erythrocyte membrane or

stroma while the third was directly associated with the

parasite [43]. It is this latter antigen, designated

"Antigen 3", which is postulated to represent merozoite

surface coat material that is shed into the culture media

when the merozoite invades an erythrocyte. Bovine antisera

to B. bovis culture supernatant reacts with the merozoite

surface and causes morphologic alterations in the surface

coat, aggregation, and lysis of merozoites [40]. By analogy

with the merozoites of Plasmodium and species of Babesia

that infect rodents, this antibody should prevent parasite

invasion of erythrocytes and mediate their immune











17

destruction by the mononuclear phagocyte system. The other

two antigens are apparently precipitated as fibrinogen-

associated complexes that in vivo may affect coagulation,

fibrinolysis, and immune recognition of parasite proteins.

Immunization with soluble antigens in supernatants of

MASP cultures has produced conflicting results. Partial

protection was induced in four 18-month-old Bos taurus

cattle that received two subcutaneous doses (in saponin

adjuvant) and were challenged three months later with 1000

B. bovis-infected Boophilus ticks [40]. Vaccinated cattle

were clinically less severely affected than identically

challenged controls.

A subsequent study showed protection of 18 yearling

heifers for four to six months after a similar vaccine

regimen and homologous challenge by intramuscular injection

of 108 infected erythrocytes [44]. Immunization induced

marked IFA serum titers that peaked 1 week after the second

vaccination and were allowed to decline to approximately

undetectable levels prior to challenge. Following

challenge, vaccinated animals developed significantly higher

IFA titers than non-vaccinated groups, and more than 50

percent of the controls died of acute babesiosis.

In both of these trials, Babesia bovis parasitemias

were observed in all challenged cattle, but in the tick













challenge experiment lower parasitemias were generally

associated with survival [40]. Although the initial

replication of challenge parasites caused a reduction in PCV

in all cattle, the destruction of host erythrocytes was

significantly less in the vaccinated animals that were

challenged by intramuscular injection [44].

The conflicting data on culture-derived immunization

has been published from Australia [45,46]. This group used

an identical vaccine protocol in yearling and 2-year-old Bos

taurus cattle that were challenged at 20, 70, or 179 days by

intravenous inoculation of 108 organisms of a heterologous

virulent strain. Whereas a single dose of a live attenuated

vaccine strain of B. bovis provided strong protection,

soluble antigens derived from MASP cultures of the same

organism induced only partial protection and 6 of 12 animals

required treatment to prevent death. All members of the

unvaccinated or saponin-recipient control groups needed to

be treated. The persistent problem of contaminating host

proteins was illustrated by the detection of bovine blood

group antibodies in 4 of the 5 cattle that received two

doses of culture-derived antigen [45]. Even soluble B.

bovis antigens harvested from heterologous rabbit MASP

culture systems are initially contaminated with 8 or 9

bovine proteins and the problem is more severe for antigens













produced in bovine systems [43].

The ability of culture-derived exoantigens to protect

against heterologous challenge has since been examined using

Latin American isolates [47,48]. Exoantigen from a

Venezuelan strain induced a significant degree of cross-

protection in cattle challenged with five different

geographic isolates, while exoantigen from a Mexico isolate

produced a low degree of cross-immunity. Vaccination did

not prevent replication of parasites but did decrease the

change in packed cell volume and increase the 6 week post-

challenge weight gain.

Soluble antigen from microaerophilous, stationary phase

cultures of B. bovis is unlikely to find use as an

acceptable, effective immunoprophylactic for bovine

babesiosis. The antigen is difficult and expensive to

manufacture in significant amounts and must ultimately be

freed of bovine contaminants. Nevertheless this work again

demonstrates the potential for nonviable parasite antigen as

a protective immunogen.


Recombinant Antigens of Babesia bovis

Little progress has been made in cloning genes from

Babesia species, and there is only one report describing the

immune response to a recombinant B. bovis antigen [49]. A













portion of the gene encoding a 44 kDa B. bovis merozoite

surface antigen was cloned into bacteriophage Lambda gt11.

This clone was selected from a genomic expression library

screened using a panel of monoclonal antibodies against

merozoite surface antigens. Cattle immunized with the

fusion protein purified from E. coli produced antibodies

that reacted with the surface of live merozoites and

immunoprecipitated a native protein of 44 kDa.

Unfortunately, the level of antibody response among

vaccinates was highly variable. The investigators are now

looking for another adjuvant or method of antigen

presentation that will produce a more consistent antibody

titer.

In another report, a positive clone was detected in a

Lambda gt11-B. bovis cDNA library screened with sera from

cattle premunized with the Australian vaccine strain of B.

bovis [50]. The cDNA insert was 235 bp long and contained

coding sequences for only 44 amino acids. Antibodies

affinity purified from bovine sera by antibody select [97]

identified a high molecular weight antigen (220 kDa) in

immunoblots of infected erythrocytes. The same antibodies

were used in an indirect immunofluorescent assay to locate

the native molecule within the parasite in infected cells.

In this study, the recombinant protein was not tested as an













immunogen.

Although several additional genes of B. bovis have been

cloned, the functions of these genes and the nature of their

gene products is not known [51-53]. The genes have been

used, however, to establish that isolates of B. bovis are

composed of heterogeneous subpopulations and to identify a

highly polymorphic gene family. The authors suggested, but

provided no evidence, that sequence diversity at the

polymorphic locus, designated BabR, played a role in

antigenic variation of the parasite. It remains to be shown

that the proteins encoded by BabR or any of these other

genes are relevant to protective immunity.



Summary

There is compelling evidence that a safe, effective

vaccine against bovine babesiosis is an achievable goal.

Recovery from natural infection, premunition by attenuated

strains, and drug cure of a significant parasitemia all

provide strong, enduring protection against clinical

disease. Passive transfer of immune sera and gamma-globulin

has been used to show this protection is mediated in large

part by antibody [54,55]. Antibodies directed against

surface antigens exposed during the short extracellular

phases of hemoparasite infections have been shown to prevent













invasion of erythrocytes and presumably opsonize parasites

for phagocytosis [56,57]. Likewise, antibodies against new

erythrocyte antigens that are encoded by hemoparasite DNA

and expressed on the surface of infected cells mediate the

removal of intracellular stages from the circulation

[56,58].

Recently, immunization with non-viable, semi-defined

parasite antigens has been shown to induce a host protective

response that significantly reduces mortality, overt

disease, and deleterious clinicopathologic changes. The

antigen-specific nature of this kind of response is

demonstrated by the effects of bovine anti-sera raised

against the three major antigens of f. bovis in supernatants

of MASP cultures [40]. These cattle are protected against

clinical disease due to a tick-transmitted homologous

challenge, and their antisera react with merozoites to cause

aggregation, lysis, and morphologic changes in the surface

coat. The failure to exploit killed antigen as an effective

vaccine reflects the complexity of the disease and the

serious problems involved in obtaining sufficient quantities

of parasite antigen that is well characterized and pure.

Many of these problems can be overcome by the application of

current molecular biology techniques.













Research Strategy

The long term goal of my work is development of a safe,

effective vaccine against Babesia bovis. As a part of that

goal, the specific aims of the work detailed in this

dissertation were:


(1) Identification and characterization of Babesia bovis

merozoite surface antigens using antibodies from cattle

protected against babesiosis.



(2) Identification and characterization of cDNA clones

expressing recombinant merozoite surface antigens in

Escherichia coli.


A prerequisite for an antigenically defined vaccine

against B. bovis is the identification of parasite antigens

which are necessary or sufficient for the induction and

expression of a host-protective immune response. Toward

that end, I identified and characterized candidate host-

protective antigens located on the surface of the merozoite

of B. bovis. These antigens were radioiodinated on live

merozoites by a surface specific method (lactoperoxidase)

[59-61] and immunoprecipitated using antibodies from cattle

proven to be protected against babesiosis.











24

Kinetic models predict that merozoite surface antigens

which are relevant to protective immunity must induce high

levels of antibody to ensure antibody-merozoite binding

during the parasite's brief extracellular phase [62]. Of

eight surface proteins identified using antibodies from

protected cattle, four were shown to be immunodominant blood

stage antigens. A 42 kilodalton (kDa) glycoprotein,

designated Bv42, elicited the highest antibody titer. I

hypothesized that the immunodominant surface proteins,

especially Bv42, were important to the induction of the

protective immune response and set out to clone the gene(s)

encoding one or more of these candidate immunogens.

Polyadenylated mRNA collected from asynchronous MASP

cultures of a cloned B. bovis isolate was used to construct

a cDNA expression library in E. coli [63-66]. cDNA was

synthesized by the Gubler/Hoffman technique, which is highly

efficient for producing full length cDNA copies of isolated

mRNA [67]. cDNA was cloned into the unique Eco RI site in

the Lac Z gene of bacteriophage Lambda ZAP II (Stratagene,

La Jolla, CA, USA).

Lambda ZAP II bacteriophage was chosen as a cloning

vector for the following reasons: 1) lambda phage is highly

efficient for in vitro packaging and introduction of DNA

into a bacterial host [69], (2) Lambda ZAP II has a multiple













cloning site which simplifies subcloning of the insert,

especially if it contains an internal Eco RI site, (3)

addition of a helper phage to Lambda ZAP II allows in vivo

excision of a circularized phagemid which contains the

insert and can be rescued for plating onto E. coli [70],

(4) DNA inserted into the Eco RI cloning site of Lambda ZAP

II is expressed as a fusion protein containing only 3.9 kDa

of vector encoded polypeptide, and (5) the E. coli host

strain carries the lac repressor gene, lac I, so that the

potentially deleterious expression of recombinant genes can

be conditionally down regulated.

Recombinant plaques were immobilized on nitrocellulose

membranes and screened for the expression of immunodominant

merozoite surface antigens using polyclonal, monospecific

rabbit sera [71]. A clone encoding the immunodominant

merozoite surface protein Bv42 was selected for further

characterization. In order to prove definitively that the

gene expressed in E. coli encoded Bv42, antibodies were

raised against the recombinant antigen in rabbits and cattle

and shown to recognize the native molecule.

A recombinant merozoite surface protein proposed for

immunization should bear surface epitopes to induce a

protective immune response against B. bovis merozoites.

Importantly, the recombinant Bv42 protein from E. coli is











26

recognized by Babb 35A4, a monoclonal antibody against

native Bv42 that reacts with the surface of live merozoites

and thus defines a surface exposed epitope [72]. This

observation indicates that the recombinant protein closely

reproduces at least one surface exposed region of native

Bv42.














CHAPTER 2

MOLECULAR CHARACTERIZATION OF BABESIA BOVIS MEROZOITE
SURFACE PROTEINS BEARING EPITOPES IMMUNODOMINANT IN
PROTECTED CATTLE



Summary

Eight surface-radioiodinated merozoite proteins from a

cloned, pathogenic isolate of Babesia bovis can be

immunoprecipitated by antibody from cattle that are

completely protected against clinical babesiosis. Among

these eight surface proteins, the 55 and 42 kDa molecules

are biosynthetically labeled with 3H-glucosamine. The 42

kDa glycoprotein can also be labeled with 3H-myristic acid

and partitions exclusively into the detergent phase in

Triton X-114 extracts, indicating that it is an integral

membrane protein and suggesting that it is anchored by a

glycosylphosphatidylinositol moiety. Antibody mediated

protection against B. bovis merozoites most likely requires

a high level of circulating antibody to ensure antibody-

merozoite binding during the parasite's brief

extraerythrocytic phase. Antibodies in diluted sera

selectively recognize the 120, 85, 55, and 42 kDa surface

proteins. Only the 42 kDa integral membrane protein is

reactive with serum antibodies diluted > 1:16,000. Thus, I













hypothesize that these immunodominant proteins, especially

the transmembrane 42 kDa glycoprotein, are important to the

induction of the protective immune response and are

candidates for an improved vaccine against babesiosis.



Introduction

Animals that survive natural field infection or that

recover from infection with an attenuated strain are

protected against clinical disease [19,74]. These

observations indicate that the host can mount a protective

immune response after exposure to parasite antigens.

Merozoite surface proteins are important in the pathogenesis

of babesiosis due to their role in the parasite's

recognition of, attachment to and penetration of host eryth-

rocytes [75]. Accessible to the immune system and present

on the disease causing stage, these antigens are potential

targets of the protective immune response and, therefore,

candidates for vaccine development. The ability of

merozoite surface proteins to elicit a protective immune

response has been shown with Plasmodium falciparum. The

precursor protein for three major merozoite surface

polypeptides can induce complete protection against a lethal

challenge in immunized monkeys [76]. Importantly, post-

translational modification of two of these P. falciparum













proteins by glycosylation may contribute to their

antigenicity [77].

Seven B. bovis merozoite surface antigens have

previously been identified using monoclonal antibodies

against surface exposed epitopes [72,49]. These monoclonal

antibodies reacted with live merozoites in indirect

immunofluoresence assays, and immunoprecipitated

radiolabeled proteins of 16, 37, 42, 44, 60, 85, and 225

kDa. In this study, additional methods are used to more

comprehensively detect merozoite surface targets and to

characterize the post-translational modification of these

proteins. Importantly, a cloned isolate of B. bovis was

examined in order to avoid confusion generated by

intraspecific antigenic variation [51,78]. Antibodies from

each of five cattle completely protected against clinical

babesiosis identify immunodominant parasite antigens that I

hypothesize are relevant to the protective immune response.


Materials and Methods

Immune animals A Mexico isolate of B. bovis was cloned in

vitro by limiting dilution as described [66]. Five 4- to 5-

month old Holstein steers were each infected by intravenous

(IV) subinoculation of 5 x 107 infected erythrocytes (iRBC)

from a splenectomized calf. The cattle were challenged













three times IV with 108 iRBC each, once by subinoculation

and twice using culture derived parasites. At 127 days post

inoculation, the five cattle and three weight-matched,

previously uninfected controls were challenged IV with 109

iRBC from culture. The packed cell volume and parasitemia

of all animals was monitored daily. Sera used in the

following experiments was collected at 126 days post

inoculation.


Microaerophilous stationary phase cultures \ Metabolic

labeling The cloned isolate was cultured in 24-well plates

or flasks by a modification of a previously described method

[39]. For metabolic labeling, parasites were grown for 12

hours in complete medium or methionine-deficient medium. At

12 hours the media were replaced with media containing 200

or 400 uCi 35S-methionine, 3H-glucosamine, or 3H-myristic

acid. Cultures were harvested at 24 hours and iRBC's were

washed five times in cold phosphate buffered saline (PBS)

prior to extraction in lysis buffer [50 mM Tris pH 8.0, 1%

(v/v) Nonidet P-40 (NP-40), 5mM EDTA, 0.1mM N-alpha-p-tosyl-

L-lysyl-choromethylketone, and 1mM phenylmethyl-

sulfonylfluoride (PMSF)]. No radiolabel was incorporated

into protein in uninfected MASP control cultures that

received radioisotope.













Merozoite Purification A high percentage of parasitized

erythrocytes was produced in vitro by progressively reducing

the packed cell volume [79]. Merozoites were purified by a

previously described technique [80]. Viability was assessed

by mixing merozoites with 6-carboxyfluorescein diacetate.

This method, previously used for demonstrating viability of

Babesia sp. [72,81], relies on an intact cell membrane to

prevent leakage of fluorescein from the cell [82].

Viability was expressed as the percentage of total parasites

that were fluorescent.


Surface radioiodination Merozoite surface proteins were

radioiodinated using lactoperoxidase as previously described

[83]. Normal red blood cells (nRBC) from uninfected control

cultures were collected, washed three times in PBS, and

radiolabeled identically. An equivalent number of nRBC

ghosts were prepared by lysing washed uninfected cells from

control cultures by freeze/thaw in liquid nitrogen. Ghosts

were washed free of hemoglobin in PBS by multiple

centrifugations at 35,000 X g, 20 min., 4C and discarding

the supernatant until it was clear. The final pellet was

resuspended in PBS and radiolabeled with 125I using

lactoperoxidase.











32

Immunoprecipitation Immune sera collected one day prior to

the final challenge experiment were adsorbed three times

with an equal volume of packed intact nRBC's and three times

with an equal volume of nRBC ghosts (prepared as described

above). Radiolabeled f. bovis lysate was ultracentrifuged,

sonicated, and filtered as described previously [83]. This

antigen was then incubated overnight at 40C with 15 ul of

bovine serum or 15 ul of a dilution of serum in veronal

buffered saline (VBS) pH 7.4, containing 1% (v/v) NP-40.

One hundred and fifty microliters of 10% (v/v) formalinized

Protein G-bearing Streptococcus (Omnisorb@; Calbiochem; San

Diego, CA) in VBS pH 7.4, 1% (v/v) NP-40, 0.1% (w/v) gelatin

were added and incubated for 2 hours at 40C [84]. The

precipitates were washed twice with VBS, 1% (v/v) NP-40;

four times with VBS, 2 M NaCl, 1% (v/v) NP-40, 10 mM EDTA;

and twice more with VBS, 1% (v/v) NP-40. The precipitated

protein was eluted by boiling in sodium dodecyl sulfate

polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer

containing 2% SDS and 2.5% 2-mercaptoethanol [83].


SDS-PAGE Immunoprecipitates were electrophoresed under

reducing conditions in 7.5 to 17.5% continuous gradient

polyacrylamide gels [83,85].













Phase separation of Triton X-114 Washed iRBC's from 35S-

methionine labeled cultures were lysed in 10 mM Tris, 154 mM

NaCl pH 7.4, 1% (v/v) Triton X-114, 1mM PMSF at 0-40C and

frozen at -200C. For protein separation, the antigen

extract was first ultracentrifuged, sonicated and filtered

as previously described [83]. The Triton X-114 solubilized

proteins (107 protein bound counts per minute in a volume of

2.0 ml) were subjected to temperature-dependent phase

separation [86]. One million protein bound counts per

minute of each phase and the starting antigen preparation

were immunoprecipitated as described above.


Immunoblotting Parasite antigen for immunoblotting was

prepared from MASP erythrocyte cultures with approximately

25% parasitized erythrocytes. Briefly, iRBC's and nRBC

controls were washed two times in cold Puck's saline G and

two times in cold PBS, resuspended in PBS, counted, and

frozen at -200C. To remove hemoglobin from lysed cells the

samples were thawed and washed in cold PBS (43,000 X g, 20

min., 40C) until the discarded supernatant was clear. The

final pellet was detergent extracted with lysis buffer and

processed as described [83]. Twenty five million iRBC's or

an equivalent total number of nRBC's (108) were mixed with













3X SDS-PAGE sample buffer, boiled for 3 minutes,

electrophoresed in a 7.5 to 17.5% continuous gradient

polyacrylamide gel, and electrophoretically transferred to a

nitrocellulose membrane [87]. The nitrocellulose was

blocked for 4-6 hours in VBS (pH 7.4) containing 0.25% (v/v)

Tween-20, 0.25% (w/v) gelatin (blocking buffer), and reacted

overnight at room temperature with immune sera diluted in

blocking buffer [84]. After washing three times in blocking

buffer and two times in VBS containing 0.1% (w/v) gelatin,

the nitrocellulose was incubated for 2 hours at room

temperature with 2.5 uCi of 125I-Protein G (Amersham Corp;

Arlington Heights, IL) diluted in VBS containing 0.1% (w/v)

gelatin. The nitrocellulose was washed twice with VBS, 0.1%

(w/v) gelatin and four times with 1M NaCl, 10 mM EDTA, 0.25%

(v/v) Tween-20, and exposed to X-ray film with an

intensifying screen at -700C.


Results

Challenge of immune animals All five cattle infected and

repeatedly challenged with a cloned isolate of B. bovis were

completely protected against a homologous challenge.

Control cattle in the final challenge experiment experienced

a 28% reduction in packed cell volume (Fig. 2-1) (p.0005

when compared to previously infected cattle; Student's













paired t test), while only the initial infection caused a

significant reduction in packed cell volume in the exper-

imental group (39%). Despite producing extensive hemolysis,

this isolate seldom produces a circulating parasitemia

greater than 0.1% in non-splenectomized cattle. Immune sera

used in the following experiments were collected one day

prior to the final challenge.



Identification and characterization of merozoite surface

proteins Merozoites purified from culture were 95-100%

viable by 6-carboxyfluorescein diacetate staining.

Immunoprecipitation of surface radioiodinated proteins with

serum antibodies from all five protected animals identified

seven dominant surface proteins with relative molecular

weights of 250, 120, 98, 85, 55, 42, and 37 kilodaltons

(Fig. 2-2A). The 250 kDa protein does not enter the

resolving gel in a standard 14 cm. 7.5-17.5% polyacrylamide

gel but is clearly resolved in a 25 cm. gel (Fig. 2-2B). An

eighth protein of 25 kDa was immunoprecipitated by serum

antibodies from two calves (lanes 7 and 10). Control

immunoprecipitation of radioiodinated intact nRBC and nRBC

ghosts revealed no specific bands on SDS-PAGE (data not

shown).













To confirm that these proteins were parasite derived,

the immune sera were used to immunoprecipitate "S-

methionine metabolically labeled parasite proteins. The

immunoprecipitated 35S antigen profile was identical in all

five protected animals. When run side by side with 35S-

methionine labeled precipitated proteins, the radioiodinated

120, 98, 85, 55 and 42 kDa surface proteins comigrate with

metabolically labeled antigens (Fig. 2-3). The 25 and 37

kDa proteins were inconsistently precipitated from 35S-

methionine labeled antigen (Fig. 2-6).

Surface protein glycosylation and myristylation were

examined by immunoprecipitation of blood stage antigen after

in vitro incorporation of 3H-glucosamine and 3H-myristic

acid. Comigration of 125I and 3H-labeled antigen in a

polyacrylamide gel shows that the 55 and 42 kDa surface

labeled proteins are glycosylated and the 42 kDa

glycoprotein is myristylated (Fig. 2-4). Neither radiolabel

was incorporated into host protein in nRBC control cultures.

The 34 kDa glycoprotein seen in lane 4 of Figure 4 cannot be

consistently surface labeled and immunoprecipitated.

Temperature dependent phase separation of Triton X-114

was used to determine whether the surface exposed antigens

were transmembrane proteins. Parasite proteins were

metabolically labeled in culture with "S-methionine and













solubilized in 1% Triton X-114 at 0-4C. The antigen

preparation was warmed above the detergent's cloud point

(200C) and separated into aqueous and detergent phases by

centrifugation. Immunoprecipitation from each phase and the

starting solution shows that the 42 kDa antigen partitions

exclusively into the detergent phase (Fig. 2-5). The other

surface proteins partition into the aqueous phase.


Immunogenicity To determine the relative immunogenicity of

these surface proteins in protected animals, antibodies in

sequentially diluted sera were used to immunoprecipitate

metabolically labeled antigen. The reactivity of these

antibodies against the majority of 3S-labeled proteins is

lost at a 1:160 serum dilution (Fig. 2-6). However the 120,

55, and 42 kDa surface proteins remain reactive at serum

dilutions of 1:160 to 1:640. Because interpretation of

immunoprecipitation is dependent on the specific

radioactivity of labeled proteins, antibodies in diluted

sera were also examined for their reactivity with parasite

antigens by immunoblotting (Fig. 2-7). Compared to

undiluted serum, antibodies in immune serum diluted 1:500

recognize a limited number of blood stage proteins,

including proteins that migrate at the same molecular weight

as the 120, 85, 55 and 42 kDa surface antigens. A protein













that comigrates with the 42 kDa integral membrane

glycoprotein is consistently recognized at serum dilutions

>1:16,000.


Discussion

The immune response to Babesia bovis infection protects

surviving cattle against babesiosis. Previous work has also

shown that nonviable parasite antigens will induce partial

protection in cattle, and supports a subunit approach to

vaccine development [27,34,40,37]. My strategy for an

improved vaccine is to identify and characterize surface

antigens of the merozoite, the blood stage that is infective

for the erythrocyte and accessible to the host immune

system. Native and recombinant parasite antigens can then

be purified, or expressed by recombinant vaccine vectors,

and evaluated as immunogens.

Seven B. bovis merozoite surface proteins have been

identified on an uncloned Mexico isolate using monoclonal

antibodies [72,49]. By indirect immunofluorescence, these

antibodies reacted with the surface of live merozoites and,

therefore, defined proteins containing regions that are

accessible to antibody binding. Four proteins of 37, 42,

44, and 85 kDa were distributed over the entire surface of

the merozoite and could be surface radioiodinated.













In the present study, antibodies in sera from cattle

that are proven to be protected against virulent challenge

also react with surface iodinated proteins of 37, 42 and 85

kDa. The 42 and 85 kDa surface proteins can be

immunoprecipitated using the previously described monoclonal

antibodies (data not shown). In addition, antibodies in

these sera identify five merozoite surface proteins of 250,

120, 98, 55, and 25 kDa not previously detected with

monoclonal antibodies. The parasite specificity of all

eight surface antigens is confirmed by the failure of serum

antibodies from protected cattle to react with iodinated

normal erythrocyte proteins and by the biosynthetic labeling

of these proteins in cultures in which only parasite mRNA is

translated [83,81]. The use of a cloned isolate in this

work is significant in that all eight surface exposed

proteins are expressed on the merozoite and do not represent

variants in subpopulations. Uncloned B. bovis isolates,

including the current Australia vaccine strain, are composed

of subpopulations that differ in antigenicity, virulence and

abundance within an isolate [51,78].

To induce immunity, a recombinant parasite antigen may

need to be presented by a eukaryotic expression vector that

can post-translationally modify the protein and preserve

native epitopes including carbohydrate moieties [73].













Carbohydrate side chains are important for the antibody

reactivity of Plasmodium falciparum antigens and are likely

to be significant in Babesia spp. [77]. By

immunoprecipitation of 3H-glucosamine labeled parasite

antigen, the 55 and 42 kDa B. bovis molecules are shown to

be glycosylated. The contribution of carbohydrate epitopes

to the immunogenicity of these proteins needs further study.

The 42 kDa merozoite surface glycoprotein was

identified as a transmembrane protein using phase separation

in Triton X-114. In this technique, integral membrane

proteins interact with the nonionic detergent via their

hydrophobic domain and when warmed above 200C can be

separated from hydrophilic proteins (which are excluded from

detergent micelles) by centrifugation [86]. The 35S-

methionine labeled 42 kDa molecule was immunoprecipitated

almost exclusively from the detergent phase. Additional

surface exposed proteins that separate into the aqueous

phase may be peripheral membrane proteins or may be

partitioning anomalously under these conditions (88,89].

Recognized causes of anomalous partitioning of integral

membrane proteins into the aqueous phase of Triton X-114

include (1) large hydrophilic moieties such as

carbohydrates, (2) the formation of oligomers which protect

hydrophobic domains, and (3) noncovalent association with











41

peripheral hydrophilicc) proteins such as those from the the

cytoskeleton (90). The strong hydrophobicity of the 42 kDa

glycoprotein may indicate that appropriate presentation,

possibly in the context of a membrane, will be required for

it to induce an effective immune response against the

merozoite.

Most integral membrane proteins are anchored in the

membrane by a hydrophobic stretch of amino acids. However,

a significant number of eukaryotic integral membrane

proteins are held in the membrane by a phosphatidylinositol

lipid moiety [91]. These proteins, which include the major

surface antigens of the parasitic protozoa Trypanosoma

brucei, Leishmania major and Plasmodium falciparum, are

covalently modified at their C-terminus with a glucosamine-

containing carbohydrate chain and the fatty acid myristate

[92-94]. By analogy with these protozoal surface proteins,

the incorporation of both 3H-glucosamine and 3H-myristic

acid into the 42 kDa protein suggests that this molecule is

embedded in the merozoite membrane by a similar glycosyl-

phosphatidylinositol anchor [95]. The previously described

T. brucei and L. maior anchors are sensitive to specific

degradation with phospholipases, releasing the protein from

the cell surface and exposing a cross-reactive epitope.

Similar studies are needed to characterize the role of













myristic acid and glycans in anchoring the B. bovis 42 kDa

protein.

In general, antibodies that inhibit P. falciparum

merozoite invasion of erythrocytes in vitro must be present

in high concentrations [62]. High antibody levels ensure

that enough surface specific antibody will be available to

effectively react with parasites during their short

extracellular phase. Cattle that are immune to infection

with a cloned isolate of Babesia bovis produce high titered

antibody that is preferentially directed against four

merozoite surface proteins with relative molecular weights

of 42, 55, 85, and 120 kDa. The 42 kDa integral membrane

glycoprotein that elicits the highest antibody titer has

previously been shown to be species specific and to contain

epitopes conserved in at least four geographic isolates [96

and unpublished data]. The preferential response of

protected animals to these four proteins, particularly the

42 kDa glycoprotein, leads me to hypothesize that these

parasite proteins are important in the induction of

protective immunity. Native and recombinant protein are

being isolated to test this hypothesis.






















0 -10-
C


0
C
(0 -20
,-3
0



-30


\ I


10 15 2
DAYS POST-INFECTION


FIGURE 2-1 Animals previously infected with the cloned
Mexico isolate of B. bovis are immune to virulent challenge.
Percentage change in PCV was measured daily in previously
infected (-), n=5, and uninfected (---), n=3, animals
following inoculation with 109 infected erythrocytes of the
cloned isolate.
















0 1 2 34 5 6 7 8 910


200-


92.5-
69-


46-


-200








- 92.5


30-

-69


14.3-






FIGURE 2-2 Antibodies in immune sera immunoprecipitate
eight B. bovis merozoite surface proteins. A. Gradient
SDS-PAGE (14 cm.). Radioiodinated merozoite proteins were
immunoprecipitated using Protein G-bearing Streptococcus
alone (Lane 0),_pre-infection sera (Lanes 1-5) or immune
serum (Lanes 6-10) from animals B101 (Lane 1,6), B102 (Lane
2,7), B108 (Lane 3,8), B109 (Lane 4,9), and B116 (Lane
5,10). B. A 250 kDa radioiodinated protein (arrow)
immunoprecipitated by antibodies in immune sera was more
clearly resolved in a 25 cm. gradient SDS-polyacrylamide
gel.


A &













1 2 3 4 5 6 7


m- W mso a

- --* ? *


i


w" V


-200



-92.5

-69



-46


- 30


-14.3


FIGURE 2-3 Surface radiolabeled B. bovis merozoite proteins
(Lanes 1,7) comigrate with metabolically labeled antigens
(Lanes 2-6). Radioiodinated merozoite surface proteins were
immunopreicipitated using immune sera form animals B102 and
B116 (Lanes 1 and 7, respectively). "S-methionine labeled
antigens were immunoprecipitated using immune sera from B101
(Lane 2), B102 (Lane 3), B108 (Lane 4), B109 (Lane 5), and
B116 (Lane 6).


4 5 6 7


1 2 3


4-0**















1 2


200-



92.5-


Q40.


A6404


FIGURE 2-4 Post-translational modification of merozoite
surface proteins. Antigen was metabolically labeled in
erythrocyte cultures with 3H- myristic acid (Lanes 1,2) or
3H-glucosamine (Lanes 3,4). Antibodies in immune serum
(animal B116) immunoprecipitated radiolabeled proteins from
infected (Lanes 1,4) but not uninfected (Lanes 2,3)
cultures. The 42 kDa myristylated B. bovis protein and the
42 and 55 kDa glycosylated proteins comigrated with
radioiodinated merozoite surface proteins (arrows) that were
immunoprecipitated using immune sera (Lane 5, B109).


69-


46-


30-


3 4












1 2 3


200-


92.5-

69-




46-



30-


-'4


14.3-


FIGURE 2-5 The 42 kDa B. bovis merozoite surface protein
partitions into the detergent phase of Triton X-114.
Following temperature dependent phase separation, immune
serum from animal B101 was used to immunoprecipitate 35S-
methionine labeled antigen from the starting 1% Triton X-114
antigen extract (lane 1), the aqueous phase (lane 2) and the
detergent phase (lane 3). Arrow indicates the 42 kDa
protein.


. -f !

















200-


92.5-

69- ----


46-



30-





14.3-







FIGURE 2-6 Immune cattle produce high titered antibody that
is preferentially directed against the 125, 55, and 42 kDa
merozoite surface proteins. "S-methionine metabolically
labeled B. bovis proteins were immunoprecipitated using
undiluted preinfection serum from animal B101 (Lane 1) and
B101 immune sera that was undiluted (Lane 2) or diluted
1:40, 1:80, 1:160, 1:320, and 1:640 (Lanes 3,4,5,6, and 7,
respectively). Arrows indicate the immunodominant 120 kDa,
55 kDa and 42 kDa proteins.


















3 4


4'


92.5-
69-


46-


30-




14.3-


FIGURE 2-7 Immune cattle produce high titered antibody that
preferentially binds the 120, 55, 42, and 85 kDa merozoite
surface proteins in immunoblots. B. bovis blood stage
antigen was transblotted and probed using preinfection serum
from animal B101 diluted 1:500 (Lane 1) and B101 immune sera
diluted 1:500, 1:4000, 1:8000, and 1:16,000 (Lanes 2,3,4,
and 5,respectively). Each lane contains 2.5 x 107 iRBC.
Arrows indicate the 120, 85, 55, and 42 kDa parasite
proteins.


1 2


-













CHAPTER 3


MOLECULAR CLONING OF THE GENE ENCODING THE IMMUNODOMINANT
BABESIA BOVIS MEROZOITE SURFACE PROTEIN Bv42 AND ITS
EXPRESSION IN ESCHERICHIA COLI.



Summary

Bv42 is the immunodominant merozoite surface antigen of

Babesia bovis and a candidate immunogen against bovine

babesiosis. A 0.9 kb cDNA fragment encoding Bv42 was cloned

into the unique Eco RI site of the bacteriophage vector

Lambda ZAP II. The 45 kDa recombinant protein expressed in

Escherichia coli accounted for greater than 95% of the

molecular weight of the native 42 kDa glycoprotein,

including the epitope(s) defined by monoclonal antibodies

that react with the surface of live merozoites. Antibodies

prepared against recombinant Bv42 immunoprecipitated the 42

kDa glycoprotein from surface labeled and metabolically

labeled B. bovis. Further, antibodies against the

recombinant protein bound to and removed the 42 kDa protein

recognized by surface reactive monoclonal antibody from a

preparation of native, 35S-methionine labeled parasite

antigen.













Introduction

Cattle that recover from infection with an attenuated

vaccine strain of Babesia bovis are protected against

clinical disease induced by arthropod-borne sporozoite

challenge [19]. Infection with such an attenuated blood

product has serious practical limitations, including

variation in infective dose and the potential transmission

of other blood borne diseases [74]. However, the observation

that an immune response to blood stages protects animals

challenged via the arthropod vector suggests strongly that a

host immune response to appropriate blood stage antigens

would be sufficient to protect cattle against natural

Babesia infections. My strategy toward the development of

an improved vaccine against Babesia bovis is to identify,

isolate, and test antigens exposed on the surface of the

merozoite. Merozoite surface proteins are accessible to the

host immune system and play a functional role in erythrocyte

invasion [75].

Surface reactive antibody that is relevant to

protection will likely be present at high levels to ensure

antibody-merozoite binding during the parasite's short

extracellular phase [65]. Using antibodies from cattle

completely protected against clinical babesiosis, I

previously identified four immunodominant merozoite surface













antigens having apparent molecular weights of 42, 55, 85,

and 125 kDa in SDS-polyacrylamide gels [98]. Bv42 is the

surface antigen that elicits the highest antibody titer.

This 42 kDa glycoprotein has a surface exposed region

defined by monoclonal antibodies [72] and contains epitopes

conserved in isolates from Mexico, Honduras, Australia and

Argentina [96 and unpublished data]. I report here the

molecular cloning and expression of the gene encoding Bv42.

Importantly, the recombinant protein expressed in

Escherichia coli (E. coli) is a faithful immunologic replica

of Bv42 that can be used to induce antibodies against the

native glycoprotein.



Materials and Methods

Parasites A cloned Mexico isolate was cultivated in long

term cultures by a previously described technique [39].

Parasite antigens were metabolically labeled in vitro with

35S-methionine or 3H-glucosamine as described [98].

Merozoites were isolated from cultures with a high

percentage of parasitized erythrocytes [80,79] and surface

radioiodinated using the lactoperoxidase technique [83,98].

Merozoite viability was greater than 95% as assessed by

retention of 6-carboxyfluoroescein diacetate [81,82].













B. bovis cDNA Expression Library Erythrocytes from

asynchronous B. bovis-infected blood cultures were washed

three times in Puck's saline G and stored frozen in liquid

nitrogen. Cells were thawed in lysis buffer containing 0.2

M NaCl, 0.2 M Tris-HCl pH 7.5, 1.5 mM MgCl2, 2% SDS (w/v),

and 200 ug/ml Proteinase K and then incubated in lysis

buffer at 460C for 2 hours [65]. The NaCI concentration of

the lysate was adjusted to 0.5 M and poly [A] RNA was

isolated by batch adsorption with oligo(dT) cellulose [65].

RNA eluted with 0.01 M Tris-HCl pH 7.5 was used to prepare a

blood stage cDNA library in Lambda ZAP II (Stratagene, La

Jolla, Ca, USA) [70] by a modified Gubler and Hoffman method

using Eco RI adapters (Pharmacia LKB, Piscataway, NJ, USA)

[67]. The cloned insert in plaque purified lambda phage was

subcloned into Bluescript SK(-) phagemid using the in vivo

excision capabilities of Lambda ZAP II [70].


Immunoscreening Plaque lifts onto isopropyl thiogalacto-

pyranoside soaked nitrocellulose were screened using

monospecific rabbit anti-Bv42 antisera (R-914) followed by
125I-Protein A and autoradiography [71,49]. Rabbit R-914 had

been immunized with native Bv42 protein immunoaffinity

purified using Babb 35A4, a previously described monoclonal











54

antibody [72]. Positive plaques were tested for reactivity

with monoclonal antibodies that recognize a Bv42 surface

exposed epitope [72,49] as well as an isotype control

monoclonal antibody and normal rabbit sera. Recombinant

phagemid excised from positive, plaque purified lambda phage

was tested for expression by a similar method using colony

lifts from transformed, ampicillin resistant E. coli (XL1-

Blue strain) [71].


Restriction Enzyme Digestion Lambda rBv42 phagemid DNA was

isolated from bacteria by anion exchange chromatography

(Qiagen Inc., Studio City, CA) and restriction enzyme

digested by standard methods [99]



Immunoblotting E. coli host strain XL1-Blue containing the

lambda rBv42 phagemid, an unrelated recombinant Anaplasma

marginale-Bluescript phagemid, or no phagemid was grown in

liquid culture to OD600=0.9-1.2. Bacteria washed three times

in phosphate-buffered saline (PBS) were lysed in PBS by

sonication and several cycles of freeze/thaw. The resulting

bacterial lysate was cleared by ultracentrifugation,

filtered, and sonicated again before freezing in aliquots at

-700C [83]. Antigen from equivalent numbers of bacteria or











55

from 2.5 X 107 B. bovis infected erythrocytes were separated

by sodium dodecyl sulfate polyacrylamide gel electrophoresis

(SDS-PAGE) and transblotted to nitrocellulose filters

[87,98]. Nitrocellulose filters were probed as described

for immunoscreening.


Antisera against recombinant Bv42 A whole lysate of E. coli

containing lambda rBv42 phagemid was prepared as described

above. Two rabbits (R-S19, R-S20) and two 4-5 month old

steers were immunized four times with a crude PBS lysate of

whole bacteria that had been disrupted by sonication and

multiple cycles of freeze/thaw and emulsified in Freund's

adjuvant. Two additional rabbits (R-S21 and R-S22) were

immunized four times with PBS soluble bacterial antigen that

had been cleared by ultracentifugation before emulsification

in Freund's adjuvant. Control anti-E. coli antibody was

produced by immunizing rabbits with a similar crude PBS

lysate of XL1-Blue without Bluescript phagemid (R-126) or

XL1-Blue with Bluescript phagemid encoding an unrelated

Anaplasma marQinale antigen (R-108).


Radioimmunoprecipitation / SDS-PAGE Radiolabeled B. bovis

antigen was processed and immunoprecipitated using Protein

A-bearing Staphylococcus aureus as previously described











56
[81,83]. The immunoprecipitates were electrophoresed in 7.5

to 17.5% gradient SDS-polyacrylamide gels and detected by

autoradiography [83].



Removal of native Bv42 from an antigen preparation by

immunoprecipitation using antibody against rBv42

"S-methionine labeled B. bovis antigen (106 protein bound

counts per minute/tube) was incubated with 10 ul of rabbit

R-S21 antisera or 5 ug of Babb 35A4 monoclonal antibody.

Following precipitation of immune complexes with protein A-

bearing Staphylococcus aureus, the supernatant antigen from

the R-S21 tube was transferred to a new tube and incubated

with another 10 ul of R-S21 antisera. Immune complexes were

again precipitated using protein A-bearing bacteria.

Antigen that remained in the supernatant from this

immunoprecipitation was transferred to a third tube and

incubated with 5 ug of Babb 35A4. Antigen recognized by

monoclonal antibody was precipitated by sequential

incubation with rabbit anti-mouse IgG antisera and protein

A-bearing Staphylococcus aureus. All immunoprecipitates

were eluted from bacteria and examined by SDS-PAGE as

described previously [81,83].













Results

Identification of a cDNA clone encoding Bv42 The B.

bovis cDNA library, containing 1.2 X 106 recombinant plaque

forming units, was screened for expression using a

monospecific rabbit antiserum (R-914) against Bv42. A

positive clone, designated Lambda rBv42, was plaque purified

and shown in plaque lifts to express a protein that reacted

with Babb 35A4, a monoclonal antibody against native Bv42

that recognizes an exposed epitope on the surface of live

merozoites [72]. Positive plaques did not react with normal

rabbit serum or isotype control monoclonal antibody.

Lambda rBv42 was subcloned into pBluescript SK(-)

phagemid in E. coli XL1-Blue using the in vivo excision

properties of Lambda ZAP II [70]. Bacteria containing the

recombinant phagemid expressed a protein recognized by R-914

and two surface reactive monoclonal antibodies against

native Bv42 (Babb 35A4 and 23.10.36; Fig. 3-1) [6,7].

Positive bacterial colonies were not recognized by normal

rabbit serum or control monoclonal antibody, and none of the

Bv42 reactive antibodies reacted with host strain bacteria

not containing the phagemid.













Size and location of insert The phagemid encoding

recombinant Bv42 contained a 0.9 kilobase (kb) DNA insert

that was excised as a single fragment with Eco RI. A double

restriction digest with Sma I and Kpn I excised a 0.96 kb

fragment that migrated just above the Eco RI fragment in

agarose gels. Sma I and Kpn I cut pBluescript SK(-) DNA

within the polylinker sequence on either side of the Eco RI

site and a non-recombinant fragment would be only 0.07 kb.

Digestion with other restriction enzymes (Hind III, Xho I,

Pvu II, Bam HI, Bgl I, and Pvu I) indicated the presence of

a 0.9 to 1.0 kb insert.



Characterization of recombinant protein (rBv42) In

immunoblots using R-914 and Babb 35A4, the recombinant B.

bovis protein expressed in E. coli migrated in SDS-

polyacrylamide gels at an apparent molecular weight of 45

kDa (Lane 6, Figs. 3-2 A and B), slightly larger than the

native glycoprotein (Lane 7, Figs. 3-2 A and B). Neither

antibody reacted with proteins produced by E. coli (XL1-Blue

host strain) without the phagemid or E. coli containing an

unrelated recombinant Bluescript phagemid. In addition, the

recombinant Bv42 protein could not be detected in

immunoblots probed with normal rabbit sera or isotype

control monoclonal antibody (Figs. 3-2 A and B, Lanes 1-5).













The relative molecular weight of recombinant Bv42 was

confirmed by SDS-PAGE after R-914 antisera and Babb 35A4

monoclonal antibody were used to immunoprecipitate from

bacterial antigen that had been metabolically labeled in

vitro with "S-methionine.


Antibodies against rBv42 recognize native Bv42 Following

immunization with rBv42, all four rabbits and one of two

cattle produced antibodies that immunoprecipitated a 42 kDa

molecule from native B. bovis antigen (Fig 3-3). The native

protein precipitated by antibodies raised against the

recombinant protein could be metabolically labeled in

infected erythrocyte cultures with "S-methionine or 3H-

glucosamine and was radioiodinated on live merozoites by a

surface specific method (Fig 3-4). With each radiolabeling

method, the precipitated protein comigrated in SDS-PAGE with

a 42 kDa parasite protein immunoprecipitated using the

merozoite surface reactive monoclonal antibody Babb 35A4

(Fig. 3-4).

Antibodies from a rabbit immunized with rBv42 were used

to deplete a B. bovis antigen preparation of the native 42

kDa surface protein defined by Babb 35A4. A single

immunoprecipitation using R-S21 sera precipitated a 42 kDa













protein from "S-methionine labeled parasite antigen (Fig.

3-5, Lane 2). A second immunoprecipitation from the

resulting supernatant using sera from the same rabbit showed

that none of the 42 kDa protein recognized by antibodies in

R-S21 sera remained (Fig. 3-5, Lane 3). A radiolabeled

protein was also not detected when the depleted supernatant

antigen was subsequently precipitated using Babb 35A4 (Fig.

3-5, Lane 4). In contrast, the same antigen preparation not

previously depleted with antibodies against rBv42 contained

a 42 kDa protein that could be immunoprecipitated using Babb

35A4 (Fig. 3-5, Lane 5).



Discussion

Bv42 is the immunodominant B. bovis merozoite surface

antigen, preferentially recognized by antibodies from cattle

protected against babesiosis [98]. This 42 kDa integral

membrane glycoprotein can be surface radiolabeled and was

originally defined by monoclonal antibody Babb 35A4 that

bound to an exposed epitope on live merozoites [72]. In

addition, Bv42 has previously been shown to be species

specific and to contain epitopes conserved in at least four

geographic isolates [96 and unpublished data]. These

features make Bv42 a strong candidate for inclusion into a

subunit vaccine against B. bovis.











61

The marked immunodominance of Bv42 and conservation of

epitopes in multiple isolates suggests also that this

protein will be a valuable diagnostic reagent. Affinity

purified rBv42 or a synthetic peptide corresponding to the

immunodominant region of the native protein would provide a

D. bovis-specific target antigen for use in an antibody

based diagnostic test [96].

f. bovis cDNA cloned into Lambda ZAP II and

subsequently subcloned into Bluescript SK(-) phagemid

encoded a recombinant parasite protein that is a faithful

immunologic replica of the native molecule. The recombinant

protein bears the Bv42 surface epitope defined by the

monoclonal antibody Babb 35A4, and antisera generated

against the recombinant protein immunoprecipitated the 42

kDa native B. bovis protein. It was important to prove

definitively that the 42 kDa protein precipitated by

antibodies raised against the recombinant was the same 42

kDa surface protein defined by Babb 35A4. Both antigens

could be metabolically labeled in vitro with 3H-glucosamine

and surface labeled on live merozoites with 125I, techniques

by which a limited number of parasite proteins are

radiolabeled [98]. More directly, however, antibodies

against rBv42 bound to and removed the glycoprotein













recognized by Babb 35A4 from a preparation of native

antigen.

Based on its molecular weight, the recombinant protein

is predicted to include greater than 95% of native Bv42. In

SDS-polyacrylamide gels, recombinant Bv42 migrated just

above the native protein at a relative molecular weight of

45 kDa. Expression of a gene cloned into the unique Eco RI

site of Lambda ZAP II should produce a fusion protein

containing 3.9 kDa of vector encoded polypeptide

(Stratagene). Since Bv42 would not be glycosylated in a

prokaryotic vector, rBv42 would likely migrate at a lower

relative molecular weight than the native glycoprotein.

Therefore, the 45 kDa rBv42 described here is predicted to

include at least 41 kDa of the 42 kDa native molecule.

Restriction enzyme mapping of isolated phagemid DNA

confirmed that the cDNA insert was cloned into the unique

Eco RI site of the vector. A 0.9 kb fragment was excised

using Eco RI and restriction enzymes that cut within the

multiple cloning site on either side of the Eco RI site. An

insert of this size should encode a protein of only about 30

kDa, but many parasite antigens migrate anomalously in SDS-

PAGE at higher molecular weights than predicted due to

highly repetitive primary sequences [100]. These repetitive

regions also tend to be immunodominant and may explain the













marked antibody response to Bv42 in cattle infected with B.

bovis [101]. The apparent discrepancy between the size of

the insert and the relative molecular weight of the

recombinant protein in SDS-PAGE will be resolved by

sequencing this gene.

The cloning of Bv42 will now enable testing of the

hypothesis that a monospecific immune response against this

merozoite surface antigen will protect cattle against

babesiosis. The recombinant protein expressed in E. coli

contains at least one epitope exposed on live merozoites and

has already been shown to induce antibody that recognizes

native Bv42. That recombinant protein is being isolated by

immunoaffinity chromatography and will be used to immunize

cattle prior to challenge with B. bovis. In addition, the

gene for Bv42 will be subcloned into vaccinia virus for

testing as an immunogen. Expression in vaccinia virus

should result in appropriate glycosylation and folding of

rBv42, thereby preserving potentially important epitopes

contributed by carbohydrate moieties or tertiary structure

[73].










Babb 35A4


23.10.36


'II


R -914


ill


FIGURE 3-1 E. coli containing the recombinant phagemid
express a protein recognized by two surface reactive
monoclonal antibodies against native Bv42 and by rabbit
anti-serum (R-914) that is monospecific for Bv42. Colony
lifts from XL-1 Blue host strain bacteria (right) or XL-l
Blue containing the recombinant phagemid (left) were probed
with three monoclonal antibodies that react with the surface
of live merozoites (Babb 35A4, 23.10.36, and 23.3.16), an
IgG2, monoclonal antibody isotype control, or R-914 anti-
serum. As a positive control, R-914 was also used to
detected B. bovis-infected erythrocytes that were spotted
onto nitrocellulose membranes at 100/1 ul, 10 /1 ul, 10'/1
ul, and 103/lul (left to right).


IgG 2a


23.3.16




























FIGURE 3-2 5. coli containing the recombinant Bv42
phagemid express a 45 kDa protein bearing the surface
exposed epitope defined by monoclonal antibody Babb 35A4.
Antigen was collected from the XLl-Blue E. coli host strain
(Lanes 1 and 4), XLl-Blue containing a recombinant
Bluescript SK(-) phagemid encoding an Anaplasma marginale
gene (Lanes 2 and 5), and XLl-Blue containing the rBv42-
Bluescript phagemid (Lanes 3 and 6). Bacterial antigen
(Lanes 1-6) or B. bovis iRBC antigen (Lane 7) were
transblotted and probed with normal rabbit sera (Fig. A,
Lanes 1-3), monospecific rabbit anti-Bv42 sera R-914 (Fig.
A, Lanes 4-7), an IgG2a isotype-control monoclonal antibody
control (Fig. B, Lanes 1-3), or Babb 35A4 (Fig. B, Lanes 4-
7).









1 2 3 4 5 6 7




















2 3 4 5 6 7










V.


-92.5

-69

-46


-30




-14.3


B.




-92.5

-69

-46


-30


-14.3
















3 4 5 6 7 8 9


-200



-92.5

-69


-46


- 6 -


-30




-14.3





FIGURE 3-3 Antibodies against the recombinant Bv42
protein immunoprecipitate a 42 kDa native protein from "S-
methionine labeled B. bovis antigen. Metabolically labeled
protein was immunoprecipitated using pre-immunization sera
(odd numbered lanes) or post-immunization sera (even
numbered lanes) from rabbits R-S19 (Lanes 3,4), R-S20 (Lanes
5,6), R-S21 (Lanes 7,8), and R-S22 (Lanes 9,10) and cattle
B-770 (Lanes 11,12) and B-783 (Lanes 13,14). Parasite
antigen was also precipitated using sera from control
rabbits immunized with lysates of E. coli host strain XL1-
Blue (R-126, Lane 2) or XL1-Blue containing an A. marginale
recombinant phagemid (R-108, Lane 1).


1 2


10 11 12 13 14












A. B.

1 2 3 4 1 2 3 4


-200 -200


-92.5 -92.5

-69
--69


-46
-30
-30


-14.3

-14.3



FIGURE 3-4 Bovine antibodies against the recombinant
Bv42 protein immunoprecipitate a 42 kDa native protein that
is metabolically labeled with 3H-glucosamine, surface
labeled on B. bovis merozoites, and co-migrates with the
surface protein defined by Babb 35A4. A. Parasite antigen
was labeled in vitro with 3H-glucosamine and
immunoprecipitated using pre-immunization sera from steer B-
783 (Lane 1), post-immunization sera from B-783 (Lane 2),
Babb 35A4 (Lane 3), or an IgG2a control monoclonal antibody
(Lane 4). B. Radioiodinated merozoite surface proteins
were immunoprecipitated using pre-immunization sera from B-
783 (Lane 1), post-immunization sera from B-783 (Lane 2), an
IgG2a control monoclonal antibody (Lane 3), and Babb 35A4
(Lane 4). Arrow indicates the 1I-labeled 42 kDa native B.
bovis protein. With both antigen preparations, identical
results were obtained using rabbit anti-rBv42 sera (data not
shown). Immunoprecipitations using sera from control
rabbits immunized with E. coli (R-126) or E. coli containing
a control phagemid (R-108) revealed no specific bands on
SDS-PAGE.











1 2 3 4 5 6


-200




-92.5


-69


-46




-30







-14.3



FIGURE 3-5 Antibodies against the recombinant Bv42
protein bind to and remove the glycoprotein recognized by
Babb 35A4 from a preparation of native antigen. 35S-
methionine labeled antigens of B. bovis contain a 42 kDa
protein that is immunoprecipitated using R-S21 rabbit anti-
rBv42 antibodies (Lane 2) or Babb 35A4 (Lane 5), but not
using R-S21 pre-immunization sera (Lane 1) or an IgGza
control monoclonal antibody (Lane 6). One
immunoprecipitation using R-S21 antibodies completely
removes the 42 kDa protein from an antigen preparation so
that it can no longer be immunoprecipitated using additional
R-S21 antibodies (Lane 3) or Babb 35A4 monoclonal antibody
(Lane 4).














CHAPTER 4


SUMMARY \ CONCLUSIONS


Merozoite surface antigens are potential targets of the

protective immune response to Babesia bovis and, therefore,

candidates for an improved vaccine. Antibody directed

against surface exposed regions of merozoite antigens is

likely to inhibit recognition and invasion of erythrocytes.

Alternatively, antibody against the merozoite surface could

opsonize the parasite for phagocytosis or, in conjunction

with complement, mediate merozoite lysis.

Using antibodies from cattle which are protected

against babesiosis, I have identified eight surface-

radiolabeled merozoite proteins from a cloned, pathogenic B.

bovis isolate. Among these eight surface proteins, the 55

and 42 kilodalton molecules were glycoproteins that could be

biosynthetically labeled with 3H-glucosamine. The 42 kDa

glycoprotein, designated Bv42, could also be labeled with
3H-myristic acid and partitioned into the detergent phase in

Triton X-114 extracts, indicating that it was an integral

membrane protein and suggesting that it was anchored by a

glycosylphosphatidylinositol moiety. This information may













be important for the presentation of Bv42 as an immunogen.

Native or recombinant protein may need to be presented in

the context of a membrane to generate antibody that reacts

with the antigen as it is expressed on the surface of the

merozoite.

Antibody mediated protection against B. bovis

merozoites most likely requires a high level of circulating

antibody to ensure antibody-merozoite binding during the

parasite's transient extracellular phase. Antibodies in

diluted sera from protected cattle selectively recognize the

120, 85, 55, and 42 kDa surface proteins. Only the 42 kDa

integral membrane glycoprotein Bv42 is reactive with serum

antibodies diluted > 1:16,000. I hypothesized that the

immunodominant proteins, especially Bv42, were important to

the induction of the protective immune response and set out

to test that hypothesis by first cloning the genes encoding

one or more of these candidate immunogens. Cloning and

expression will provide parasite antigen in sufficient

quantity and purity to test in immunization trials.

A 0.9 kb cDNA fragment encoding the 42 kDa

immunodominant merozoite surface antigen Bv42 was cloned

into the unique Eco RI site of the bacteriophage vector

Lambda ZAP II. Expression of recombinant DNA in the

multiple cloning site of the lac Z gene of Lambda ZAP II













should produce a fusion protein containing only 3.9 kDa of

vector encoded polypeptide. The 45 kDa recombinant B. bovis

protein expressed in Escherichia coli accounted for greater

than 95% of the molecular weight of the native 42 kDa

glycoprotein and, importantly, included the epitope(s)

defined by monoclonal antibodies that react with the surface

of live merozoites.

Antibodies were prepared against the recombinant

Babesia protein to prove definitively that the cloned gene

encoded Bv42 and to show that antibodies directed against

the recombinant protein recognize the native parasite

protein. This second point, that the recombinant protein

functions as an immunologic replica of the native protein,

is critical if recombinant Bv42 (rBv42) is to be tested as

an immunogen for babesiosis. Antibodies from animals

immunized with rBv42 immunoprecipitated the 42 kDa

glycoprotein from surface labeled and metabolically labeled

B. bovis. Further, antibodies against the recombinant

protein bound to and removed the 42 kDa protein recognized

by surface reactive monoclonal antibody Babb 35A4 from a

preparation of native, methionine labeled parasite antigen.

A successful immunization trial using an isolated B.

bovis antigen would focus the efforts to develop an

improved vaccine. In addition, however, the work to isolate











73
the genes for B. bovis antigens and develop monospecific

reagents will help answer important questions about the

basic biology of Babesia spp. Those questions will include

the biologic function of merozoite surface antigens, the

extent of antigenic diversity among different geographic

isolates, the role, if any, of antigenic variation within a

single clone of parasites, and the genetic mechanisms

involved in the expression and modification of these

important parasite molecules.














REFERENCES


1. National Research Council. (1981) Priorities in
biotechnology research for international development.
Proceedings of a workshop. pp. 1-9. National Academy
Press, Washington, D.C.

2. McCosker, D.J. (1981) The global importance of
babesiosis. In: Babesiosis. (Ristic, M. and Kreir,
J.P., eds.), pp. 1-24, Academic Press, New York.

3. Furie, W.S. (1982) Bovine babesiosis. Comp. Cont. Ed.
4, S272-S277.

4. Norvall, R.A.I., Fivaz, B.H., Lawrence, J.A. and
Daillecourt, T. (1983) Epidemiology of tick-borne
diseases of cattle in Zimbabwe. I. Babesiosis. Trop.
Anim. Hlth. Prod. 15, 87-94.

5. Young, A.S. and Morzaria, S.P. (1986) Biology of
babesia. Parasitol. Today 2, 211-219.

6. Rudzinska, M.A. (1981) Morphologic aspects of host
cell-parasite relationships in babesiosis. In:
Babesiosis. (Ristic, M. and Kreir, J.P., eds.), pp. 87-
141, Academic Press, New York.

7. Friedhoff, K.T. (1981) Morphologic aspects of babesia
in the tick. In Babesiosis. (Ristic, M. and Kreir,
J.P., eds.), pp. 143-169, Academic Press, New York.

8. Kahl, L.P., Anders, R.F., Rodwell, B.J., Timms, P. and
Mitchell, G.F. (1982) Variable and common antigens of
Babesia bovis parasites differing in strain and
virulence. J. Immunol. 129, 1700-1705.

9. Kahl, L.P., Mitchell, G.F., Dalgliesh, R.J., Stewart,
N.P., Rodwell, B.J., Mellors, L.T., Timms, P. and
Callow, L.L. (1985) Babesia bovis: Proteins of virulent
and avirulent parasites passage through ticks and
splenectomized or intact calves. Exp. Parasitol. 56,
222-235.













10. Ristic, M. (1982) An overview of bovine babesiosis with
emphasis on infection and immunity. In: Proceedings of
the Seventh National Anaplasmosis Conference. pp. 3-27.

11. Purnell, R.E. (1981) Babesiosis in various hosts. In:
Babesiosis. (Ristic, M. and Kreir, J.P., eds.), pp. 25-
63, Academic Press, New York.

12. Callow, L.L., McGregor, W., Parker, R.J. and Dalgliesh,
R.J. (1974) Immunity of cattle to Babesia argentina
after drug sterilization of infections of varying
duration. Aust. Vet. J. 50, 6-11.

13. Mahoney, D.F., Wright, I.G. and Mirre, G.B. (1973)
Bovine babesiosis: The persistence of immunity to
Babesia argentina and B. bigemina in calves (Bos
taurus) after naturally acquired infection. Ann. Trop.
Med. Parasit. 67, 197-203.

14. Tatchell, R.J. (1987) Tick control in the context of
East Cost Fever immunization. Parasitol. Today 3, 7-10.

15. Carson, C.A. and Phillips, R.S. (1981) Immunologic
response of the vertebrate host to babesia. In:
Babesiosis (Ristic, M. and Kreir, J.P. eds.), pp. 411-
413, Academic Press, New York.

16. Molecular vaccines against parasites. (1986) Parasitol.
Today 2, S11-S12.

17. Callow, L.T. (1977) Vaccination against bovine
babesiosis. Adv. Exp. Med. Biol. 93, 121-149.

18. Callow, L.L., Mellors, L.T. and McGregor, W. (1979)
Reduction in virulence of Babesia bovis due to rapid
passage in splenectomized cattle. Int. J. Parasitol. 9,
333-338.

19. Mahoney, D.F., Wright, I.G. and Goodger, B.V. (1979)
Immunity in cattle to Babesia bovis after single
infections with parasites of various origins. Aust.
Vet. J. 55, 10-12.

20. Pipano, E. and Hadani, A. (1984) In: Malaria and
Babesiosis (Ristic M., Ambroise T.D. and Kreir J.P.
eds.), pp. 263-303, Martinus Nijhoff Pub., The
Netherlands.













21. Kuttler, K.L., Zaugg, J.L. and Yunker, C.E. (1988) The
pathogenicity and immunologic relationship of a
virulent and a tissue-culture-adapted Babesia bovis.
Vet. Parasitol. 27, 239-244.

22. Buening, G.M., Kuttler, K.L. and Rodriguez, S.D. (1986)
Evaluation of a cloned Babesia bovis organism as a live
immunogen. Vet. Parasitol. 22, 235-242.

23. Pipano, E. (1980) In: Proc 11th Cong on Diseases of
Cattle (Mayer E., ed.), pp. 678-681, Bergman Press,
Haifa P.O.B.

24. Lora, C.A. (1981) Methods of immunization against
bovine babesiosis used in Latin America. In: Babesiosis
(Ristic, M. and Kreir, J.P., eds.), pp. 567-571,
Academic Press, New York.

25. Pipano, E., Weisman, Y., Raz, A. and Klinger, I. (1972)
Immunity to Babesia bigemina in calves after successful
babesicidal treatment of a previous infection. Ref.
Vet. 29, 1-8.

26. Kuttler, K.L. (1981) Chemotherapy of babesiosis: A
review. In: Babesiosis (Ristic M. and Kreir J.P., eds.)
pp. 65-85, Academic Press, New York.

27. Mahoney, D.F. (1967) Bovine babesiosis: The
immunization of cattle with killed Babesia argentina.
Exp. Parasitol. 20, 125-129.

28. Mahoney, D.F. and Wright, I.G. (1976) Babesia
argentina: Immunization of cattle with a killed antigen
against infection with a heterologous strain. Vet.
Parasitol. 2, 273-282.

29. Goodger, B.V., Wright, I.G., Mahoney, D.F. and McKenna,
R.V. (1980) Babesia bovis (argentina): Studies on the
composition and location of antigen associated with
infected erythrocytes. Int. J. Parasitol. 10, 33-36.

30. Mahoney, D.F., Wright, I.G. and Goodger, B.V. (1981)
Bovine babesiosis: The immunization of cattle with
fractions of erythrocytes infected with Babesia bovis.
Vet. Immunol. Immunopathol. 2, 145-156.













31. Goodger, B.V. and Wright, I.G. (1883) Babesial antigens
in a saline eluate of sucrose washed bovine
erythrocytes infected with Babesia bovis. Experimentia
39, 1392-1393.

32. Wright, I.G., Goodger, B.V., Rode-Bramanis, K.,
Mattick, J.S., Mahoney, D.F. and Waltisbuhl, D.F.
(1983) The characterization of an esterase derived from
Babesia bovis and its use as a vaccine. Z. Parasitenkd.
69, 703-714.

33. Goodger, B.V., Wright, I.G. and Waltisbuhl, D.H. (1983)
The lysate from bovine erythrocytes infected with
Babesia bovis. Analysis of antigens and a report on
their immunogenicity when polymerized with
glutaraldehyde. Z. Parasitenkd. 69, 473-482.

34. Goodger, B.V., Commins, M.A., Wright, I.G. and Mirre,
G.B. (1984) Babesia bovis: Vaccination of cattle
against heterologous challenge with fractions of lysate
from infected erythrocytes. Z. Parasitenkd. 70, 321-
329.

35. Goodger, B.V., Wright, I.G., Waltisbuhl, D.J. and
Mirre, G.B. (1985) Babesia bovis: Successful
vaccination against homologous challenge in
splenectomized calves using a fraction of
hemagglutinating antigen. Int. J. Parasitol. 15, 175-
179.

36. Commins, M.A., Goodger, B.V. and Wright I.G. (1985)
Proteinase in the lysate of bovine erythrocytes
infected with Babesia bovis: Initial vaccination
studies. Int. J. Parasitol. 15, 491-495.

37. Wright, I.G., White, M., Tracey-Patte, P.D., Donaldson,
R.A., Goodger, B.V., Waltisbuhl, D.J. and Mahoney, D.F.
(1985) Babesia bovis: Isolation of a protective antigen
by using monoclonal antibodies. Infect. Immun. 41, 244-
250.

38. Wright, I.G., Mirre, G.B., Rode-Bramanis, K.,
Chamerlain, M., Goodger, B.V. and Waltisbuhl, D.J.
(1985) Protective vaccination against virulent Babesia
bovis with a low molecular weight antigen. Infect.
Immun. 48, 109-112.













39. Levy, M.G. and Ristic M. (1980) Babesia bovis:
Continuous cultivation in a microaerophilous stationary
phase culture. Science 207, 1218-1220.

40. Smith, R.D., James, M.A., Ristic, M., Aikawa, M. and
Vega y Murguia, C.A. (1981) Bovine babesiosis:
Protection of cattle with culture-derived soluble
Babesia bovis antigen. Science 212, 335-338.

41. James, M.A., Levy, M.G. and Ristic M. (1981) Isolation
and partial characterization of culture-derived soluble
Babesia bovis antigens. Infect. Immun. 31, 358-361.

42. Montenegro-James, S., James, M.A. and Ristic M. (1983)
Localization of culture-derived soluble Babesia bovis
antigens in the infected erythrocyte. Vet. Parasitol.
13, 311-316.

43. James, M.A. (1984) An update of the isolation and
characterization of culture-derived soluble antigens of
Babesia bovis. Vet. Parasitol. 14, 231-237.

44. Kuttler, K.L., Levy, M.G. and Ristic, M. (1983) Cell
culture-derived Babesia bovis vaccine: Sequential
challenge exposure of protective immunity during a 6-
month postvaccination period. Am. J. Vet. Res. 44,
1456-1459.

45. Timms, P.R., Dalgliesh, R.J., Barry, D.N., Dimmock,
C.K. and Rodwell, B.J. (1983) Babesia bovis: A
comparison of culture-derived parasites, non-living
antigen and conventional vaccine in the protection of
cattle against heterologous challenge. Aust. Vet. J.
60, 75-77.

46. Timms, P., Stewart, N.P., Rodwell, B.J. and Barry, P.N.
(1984) Immune response of cattle following vaccination
with living and non-living Babesia bovis antigens. Vet.
Parasitol. 16, 243-251.

47. Montenegro-James, S., Ristic, M., Bentez, M.T., Leon,
E. and Lopez, R. (1985) Heterologous strain immunity in
bovine babesiosis using a culture-derived soluble
Babesia bovis immunogen. Vet. Parasitol. 18, 321-337.













48. Montenegro-James, S., Benitez, M.T., Leon, E., Lopez,
R. and Ristic, M. (1987) Bovine babesiosis: induction
of protective immunity with culture-derived Babesia
bovis and Babesia biqemina immunogens. Parasitol. Res.
74, 142-150.

49. Reducker, D.W., Jasmer, D.P., Goff, W.L., Perryman,
L.E., Davis, W.C. and McGuire, T.C. (1989) A
recombinant surface protein of Babesia bovis elicits
bovine antibodies that react with live merozoites. Mol.
Biochem. Parasitol. 35, 239-248.

50. Gill, A., Timms, P. and Kemp, D.J. (1987) cDNA clone
encoding a high molecular weight antigen of Babesia
bovis. Mol. Biochem. Parasitol. 22, 195-202.

51. Cowman AF, Timms P, Kemp DJ. (1981) DNA polymorphisms
and subpopulations in Babesia bovis. Mol Biochem
Parasitol 11:91-103.

52. Cowman, A.F., Bernard, 0., Stewart. N. and Kemp, D.J.
(1984) Genes of the protozoan parasite Babesia bovis
that rearrange to produce RNA species with different
sequences. Cell 37, 653-660.

53. Kemp, D.J., Easton, K.E. and Cowman, A.F. (1984)
Accurate transcription of a cloned gene from Babesia
bovis in Saccharomyces cerevisiae. Mol. Biochem.
Parasitol. 12, 61-67.

54. Mahoney, D.F. (1967) Bovine babesiosis: The passive
immunization of calves against Babesia argentina with
special reference to the role of complement fixing
antibodies. Exp. Parasitol. 20, 119-124.

55. Mahoney, D.F., Kerr, J.D., Goodger, B.V. and Wright,
I.G. (1979) The immune response of cattle to Babesia
bovis. Studies on the nature and specificity of
protection. Int. J. Parasitol. 9, 297-306.

56. Deans, J. and Cohen, S. (1983) Immunology of malaria.
Ann. Rev. Microbiol. 37, 25-49.

57. Gravely, S.M., Smith, R.D., Erp, E.E., Canto, G.J.,
Aikawa, M., Osborno, B.M. and Ristic, M. (1979) Bovine
babesiosis: Partial purification and characterization
of blood culture-derived Babesia bovis. Int. J.
Parasitol. 9, 591-598. (1979)













58. Howard, R.J. and Barnwell, J.W. (1984) Roles of surface
antigens on malaria infected red blood cells in evasion
of immunity. Contemp. Top. Immunbiol. 12, 127-200.

59. Barbet, A.F. (1982) Identification and analysis of
parasite surface antigens and parasite-induced antigens
on host cells. Vet. Parasitol. 10, 181-189.

60. Howard, R.J.J., Barnwell, W., Kao, V., Daniel, W.A. and
Aley, S.B. (1982) Radioiodination of new protein
antigens on the surface of Plasmodium knowlesi
schizont-infected erythrocytes. Mol. Biochem.
Parasitol. 6, 343-364.

61. Phillips, D.R. and Morrison, M. (1971) Exposed protein
on the intact human erythrocyte. Biochem. 10, 1766-
1771.

62. Saul, A. (1987) Kinetic constraints on the development
of a malaria vaccine. Parasite Immunol. 9, 1-9.

63. Aviv, H. and Leder, P. (1972) Purification of
biologically active globin messenger RNA by
chromatography on oligothymidylic acid-celloulose.
Proc. Natl. Acad. Sci. USA 69, 1408-1412.

64. Taylor, J.M. (1979) The isolation of eukaryotic
messenger RNA. Ann. Rev. Biochem. 48, 681-717.

65. Bradley, J.E., Bishop, G.A., St. John, T. and
Frelinger, J.A. (1988) A simple, rapid method for the
purification of Poly A+ RNA. BioTechniques 6, 114-116.

66. Rodriquez, S.D., Buening, G.M., Green, T.J. and Carson,
C.A. (1983) Cloning of Babesia bovis by in vitro
cultivation. Infect. Immun. 42, 15-18.

67. Gubler, U. and Hoffman, B. (1983) A simple and very
efficient method for generating cDNA libraries. Gene
25, 263-269.

68. Watson, C.J. and Jackson, J.F. (1985) An alternative
procedure for the synthesis of double-stranded cDNA for
cloning in phage and plasmid vectors. In: DNA cloning.
Vol I. A practical approach (Glover, D.M., ed), pp. 79-
88, IRL Press, Oxford.













69. Huynh, T.V., Young, R.A. and Davis, R.W. (1985)
Construction and screening cDNA libraries in Lambda
gtl0 and Lambda gtll. In: DNA cloning. Vol I. A
practical approach. (Glover, D.M., ed), pp. 49-78, IRL
Press, Oxford.

70. Short, J.M., Fernandez, J.M., Sorge, J.A. and Huse,
W.D. (1988) Lambda ZAP: a bacteriophage with in vivo
excision properties. Nuc. Ac. Res. 16, 7583-7600.

71. Young, R.A. and Davis, R.W. (1983) Efficient isolation
of genes using antibody probes. Proc Natl Acad Sci USA
80, 1194-1198.

72. Goff, W.L., Davis, W.C., Palmer, G.H., McElwain, T.F.,
Johnson, W.C., Bailey, J.F. and McGuire, T.C. (1988)
Identification of Babesia bovis merozoites surface
antigens by using immune sera and monoclonal
antibodies. Infect. Immun. 56, 2363-2368.

73. Moss, B. and Flexner, C. (1987) Vaccinia virus
expression vectors. Annu. Rev. Immunol. 5, 305-324.

74. Callow, L.L. (1977) Vaccination against bovine
babesiosis. In: Immunity to Blood Parasites of Animals
and Man (Miller, L.H., Pino, J.A., McKelvey, J.J.,
ed.), pp. 121-149, Plenum Press, New York.

75. Jack, R.M. and Ward, P.A. (1981) Mechanisms of entry of
Plasmodium and Babesia into red cells. In: Babesiosis
(Ristic, M. and Kreier J.P., ed.), pp. 445-458,
Academic Press, New York.

76. Siddiqui, W.A., Tam, L.Q., Kramer, K.J., Hui, G.S.N.,
Case, S.E., Yamaga, K.M., Chang, S.P., Chan E.B.T. and
Kan, S.-C. (1987) Merozoite surface coat precursor
protein completely protects Aotus monkeys against
Plasmodium falciparum malaria. Proc. Natl. Acad. Sci.
USA 84, 3014-3018.

77. Ramasamy, R. and Reese, R.T. (1986) Terminal galactose
residues and the antigenicity of Plasmodium falciparum
glycoproteins. Mol. Biochem. Parasitol. 19, 91-101.

78. Gill, A.C., Cowman, A.F., Stewart, N.P., Kemp, D.J. and
Timms, P. (1987) Babesia bovis: molecular and
biological characteristics of cloned parasite lines.
Exp. Parsitol. 63, 180-188.














79. Goff, W.L. and Yunker, C.E. (1986) Babesia bovis:
increased percentage parasitized erythrocytes in
culture and assessment of growth by incorporation of
[3H]-hypoxanthine. Exp. Parasitol. 62, 202-210.

80. Rodriquez, S.D., Buening, G.M., Vega, C.A. and Carson,
C.A. (1986) Babesia bovis: purification and
concentration of merozoites and infected bovine
erythrocytes. Exp. Parasitol. 61, 236-246.

81. McElwain, T.F., Perryman, L.E., Davis, W.C. and
McGuire, T.C. (1987) Antibodies define multiple
proteins with epitopes exposed on the surface of live
Babesia biqemina merozoites. J. Immunol. 138, 2298-
2304.

82. Rotman, B. and Papermaster, B.W. (1966) Membrane
properties of living mammalian cells as studied by
enzymatic hydrolysis of fluorogenic esters. Proc. Natl.
Acad. Sci. USA 55, 134-13 .

83. Palmer, G.H. and McGuire, T.C. (1984) Immune serum
against Anaplasma marginale initial bodies neutralizes
infectivity for cattle. J. Immunol. 133, 1010-1015.

84. Akerstrom, B., Brodin, T., Reis, K. and Bjorck, L.
(1985) Protein G: a powerful tool for binding and
detection of monoclonal and polyclonal antibodies. J.
Immunol. 135, 2589-2592.

85. Takac, B. (1979) Electrophoresis of proteins in
polyacrylamide gels. In: Immunological Methods
(Lefkovitz, T. and Persin, B., ed.), p.81, Academic
Press, New York.

86. Bordier, C. (1981) Phase separation of integral
membrane proteins in Triton X-114 solution. J. Biol.
Chem. 256, 1604-1607.

87. Towbin, H. and Gordon, H. (1984) Immunoblotting and dot
immunoblotting- current status and outlook. J. Immunol.
Methods 72, 313-340.

88. Maher, P.A. and Singer, S.J. (1985) Anomalous
interaction of the acetylcholine receptor protein with
the nonionic detergent Triton X-114. Proc. Natl. Acad.
Sci. 82, 958-962.













89. Clemetson, K.J., Bienz, D., Zahmo, M.-L. and Luscher,
E.F. (1984) Distribution of platelet glycoproteins and
phosphoproteins in hydrophobic and hydrophilic phase in
Triton X-114 phase partition. Biochim. Biophys. Acta.
778, 463-469.

90. Pryde, J.G. (1986) Triton X-114: a detergent that has
come in from the cold. Trends Biochem. Sci. 11:160-163.

91. Low, M.G. and Saltid A.R. (1987) Structural and
functional roles of glycosylphosphatidylinositol in
membranes. Science 239: 268-275.

92. Bordier, C., Etges, R.S., Ward, J., Turner, M.J. and
Cardoso de Almeida, M.L. (1986) Leishmania and
Trypanosoma surface glycoproteins have a common
glycophospholipid membrane anchor. Proc. Natl. Acad.
Aci. USA 83: 5988-5991.

93. Haldar, K., Ferguson, M.A.J. and Cross, G.A.M. (1985)
Acylation of a Plasmodium falciparum merozoite surface
antigen via sn-1,2-diacyl glycerol. J. Biol. Chem. 260,
4969-4974.

94. Braun-Breton, C., Rosenberry T.L. and Pereirz da Silva,
L. (1988) Induction of the proteolytic activity of a
membrane protein in Plasmodium falciparum by
phophatidylinositol-specific phospholipase C. Nature
332, 457-459.

95. Smythe, J.A., Coppell, R.L., Brown, G.V., Ramasamy, R.,
Kemp, D.J. and Anders, R.F. (1988) Identification of
two integral membrane proteins of Plasmodium
falciparum. Proc. Natl. Acad. Sci. 85: 5195-5199.

96. McElwain, T.F., Palmer, G.H., Goff, W.L. and McGuire,
T.C. (1988) Identification of Babesia biaemina and
Babesia bovis merozoite proteins with isolate- and
species-comon epitopes recognized by antibodies in
bovine immune sera. Infect. Immun. 56, 1658-1660.

97. Crewther, P.E., Bianco, A.E., Brown, G.V., Copperl,
R.L., Stahl, H.D., Kemp, D.J. and Anders, R.F. (1986)
Affinity purification of human antibodies directed
against cloned antigens of Plasmodium falciparum. J..
Immunol. Methods 86, 257-264.













98. Hines, S.A., McElwain, T.F., Buening, G.B. and Palmer,
G.H. (1989) Molecular characterization of Babesia bovis
merozoite surface proteins bearing epitopes
immumodominant in protected cattle. Mol. Biochem.
Parasitol, (In Press).

99. Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982)
Molecular cloning a laboratory manual. Cold Spring
Harbor Laboratories, Cold Spring Harbor, New York, 545
pp.
100. Kemp, D.J., Coppel, R.L. and Anders, R.F. (1987)
Repetitive proteins and genes of malaria. Ann. Rev.
Microbiol. 41, 181-208.

101. Zavala,F., Cochrane, A.H., Nardin, E.H., Nussenzweig,
R.S. and Nussenzwieg, V. (1983) Circumsporozoite
proteins of malaria parasites contain a single
immunodominant region with two or more identical
epitopes. J. Exp. Med. 157, 1947-1957.














BIOGRAPHICAL SKETCH


Stephen Alan Hines was born to Della Maxene Hanna Hines

and Earl Thomas Hines on March 5, 1955, in Bowling Green,

Kentucky. He received a Bachelor of Arts in May, 1977, from

Miami University in Oxford, Ohio. On June 12, 1981, Steve

graduated as a Doctor of Veterinary Medicine from Ohio State

University, College of Veterinary Medicine. The following

day he married Dr. Melissa Gail Trogdon, also a

veterinarian. At that time Steve inherited Riley, a once-

in-a-lifetime dog.

After a year of dairy and small animal practice, Steve

moved with Melissa to Gainesville, Florida, to start an

anatomic pathology residency at the University of Florida's

Veterinary Medical Teaching Hospital. With his residency

completed in July, 1984, Steve entered graduate school in

the College of Medicine, University of Florida. In

September, 1985, Steve passed his veterinary pathology

specialty boards. Six months later, he changed major

advisors to work on the immunology and molecular biology of

veterinary hemoparasites, in particular Babesia bovis, with

Guy Palmer. When Guy left Florida in May, 1988, to be











86

closer to mountains, Steve moved to Terry McElwain's

laboratory to continue his dissertation work on Babesia. In

June, 1989, Terry also left Florida to return to the same

place Guy went. Soon Steve and Melissa will be following

Terry and Guy to the mountains to faculty positions at

Washington State University, that is. It's been great being

a gator, now let's try being a cougar!

After we're gone, try to preserve the natural beauty of

Florida. It's worth fighting for.










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.


Terry F. McElwain, Chair
Assistant 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.


Lindsey"utt-Fletcher, Cochair
Associate Professor of Pathology
and Laboratory 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


Guy H. Palmer
Assistant 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.


Anthon F. Barbet
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 scope and quality, as a dissertation for
the degree of Doctor of Philosophy.

Edward Wakeland
Associate Professor of Pathology
and Laboratory Medicine




This dissertation was submitted to the Graduate Faculty of
the College of Medicine and to the Graduate School and was
accepted as partial fulfillment of the requirements for the
degree of Doctor of Philosophy.


December, 1989 ___^_ _
Dean, College of Medicine



Dean, Graduate School











































UNIVERSITY OF FLORIDA
I111111111111111111111111111111 111II11 1IIII11111
3 1262 08554 3311