An immunological and olfactory study of horn flies, "Haematobia irritans" (L.) : identification of antigens from the sal...


Material Information

An immunological and olfactory study of horn flies, "Haematobia irritans" (L.) : identification of antigens from the salivary glands and their relationships to host interaction
Physical Description:
vii, 129 leaves : ill. ; 29 cm.
Okine, James Spencer, 1958-
Publication Date:


Subjects / Keywords:
Entomology and Nematology thesis, Ph. D
Dissertations, Academic -- Entomology and Nematology -- UF
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1994.
Includes bibliographical references (leaves 110-128).
Additional Physical Form:
Also available online.
General Note:
General Note:
Statement of Responsibility:
by James Spencer Okine.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 021609120
oclc - 33394039
System ID:

Full Text








I wish to express my sincere appreciation to my major

professor and committee chairman, Dr. J. F. Butler, for the

encouragement, ideas and financial support offered me,

without which this work could never have been done. I am

grateful for the technical and moral support given to me by

Drs. S. G. Zam, J. L. Nation and H. L. Cromroy. Special

thanks go to Dr. Zam for introducing me to the field of

immunology and for technical advice and for the use of his

lab. Also sincere appreciation goes for Dr Cromroy for

making it possible for me to attend the University of

Florida and to Dr Nation, special thanks for his

understanding and kind help.

I am thankful for my wife Grace and my children Victor

Ayittey, Earl Ayiquaye and Marilyn Okailey for the

encouragement and support and most importantly for

tolerating me, especially during the last stages of my


I am deeply indebted to my parents James and Victoria

Okine for their discipline, love, support, and most

importantly the seed for learning sown in me without which I

would not have made it this far in life.

Sincere thanks go to Diana Simon who has become like

family to me, and to Robert Stewart, Margo Duncun, Jason

Byrd, Scott Mize, Haze Brown, Debbie Boyd, Dr. E. Wozniak,

Nick Hosletter and all the faculty and students of the

department for making life bearable.



ACKNOWLEDGEMENTS .......................................... ii




Importance Of The Horn Fly .......................... 1
Control Strategies For The Horn Fly................. 5
Host-Parasite Interaction ...........................12
Arthropod Defence Mechanisms ........................ 17
Vertebrate Immune System And Host-Immune Response...20
Host Location By Arthropods ......................... 25
Role Of Saliva In Blood-Feeding By Arthropods.......29
Immunohistochemistry And Antibodies As
Investigative Tools ............................ 31
Monoclonal Antibodies ...............................35
ELISA (Enzyme-Linked Immunosorbent Assay) ...........35
Research objectives ................................. 36


Introduction ........................................ 39
Materials And Methods ............................... 41
Results ............................................. 50
Discussion .......................................... 51


Introduction ........................................ 61
Materials And Methods................................ 64
Horn Fly Rearing And Culture Procedures............. 64
Dissection Of Horn Fly Salivary Glands For Antigen
Production..................................... .66
Polyclonal Antigen Production ....................... 67

DOT-Enzyme Linked Immunosorbent Assay (DOT-ELISA)
For Determining Optimum Antigen Titer..........68
Antibody Titer Determination Using Enzyme-Linked
Imunosorbent Assay (ELISA) ..................... 69
Protein Separation With Polyacrylamide Gel
Electrophoresis ................................ 71
Western Blot ........................................ 75
Immunodetection Of Salivary Gland Antigens.......... 75
Evaluation Of SGE For Carbohydrate Epitopes......... 77
Results ............................................. 78
Optimum Antigen And Antibody Titers ................. 78
Polyacrylamide Gel Electrophoresis, Western Blot
And Characterization Of Salivary Gland
Antigens....................................... 79
Periodate Oxidation ................................. 80
Discussion .......................................... 80


Introduction ........................................ 95
Materials And Methods ............................... 96
Results ............................................. 98
Discussion .......................................... 99

5 SUMMARY AND CONCLUSIONS ............................... 107

LITERATURE CITED.................................... .110

BIOGRAPHICAL SKETCH ................................. 129

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



James Spencer Okine


Chairperson: Dr Jerry Frank Butler
Major Department: Entomology and Nematology

Studies of horn fly, Haematobia irritans, salivary

antigen(s) and its response to hexane washes of repellent

and susceptible cows were conducted in relation to control.

Polyacrylamide gel electrophoresis (PAGE) analysis of

salivary gland extract (SGE) of the horn fly showed complex

proteins ranging from about 75 KDa to less than 14 KDa.

Antibodies generated in rabbits against salivary gland

extract (SGE) by direct immunization and that generated in

mice against salivary secretions of horn flies by chronic

infestation were used to immunoprecipitate protein bands

from the SDS-PAGE separated and immunoblotted SGE. The

mouse antisera precipitated two band at about 27 KDa. The


rabbit antisera precipitated about seven protein bands, one

just below 30 KDa, two band around 21.5 KDa, three between

46 and 69 KDa and a band below 200 KDa. Antisera from both

model animals detected protein bands at around 27 KDa. The

antigen at 27 KDa was therefore identified as a secretary

protein that can have immunological importance. Use of the

two types of polyclonal antibodies generated to probe

histological sections of the head and thorax of the horn fly

showed that the rabbit antisera specifically localized the

epithelial cells making up the salivary glands. The mouse

antisera produced nonspecific binding when compared to

normal mouse serum.

Periodate oxidation test showed that the SGE possesses

a carbohydrate epitope. The optical density of periodate

treated SGE was reduced by a greater percentage when reacted

with polyclonal antibodies from the rabbit than from the


Use of the ten-port olfactometer for testing hexane

washes of susceptible and attractant bovines and selected

natural chemicals supplied by International flavors and

fragrances (IFF) showed the olfactometer to be a reliable

tool for testing efficacy of chemicals against horn flies.

Also, the result showed that the individual host animal

phenomenon of resistance and attractancy to horn flies is

real suggesting further research to identify the chemicals

produced by the host which produces the repellent activity.



Importance Of The Horn Fly

About half of the annual agricultural income in the

United States of America comes from livestock (Kunz et al.,

1991). USA livestock farming is considered one of the most

efficient and productive in the world, partly because of

modern methods used to control parasites of these animals.

Despite great success, there remains a need to control some

ectoparasites and replace failing methods to prevent further

losses and to provide for more efficient production. Most

discussion of ectoparasites dwells on their role as vectors

of disease, but the direct costs of hematophagous

ectoparasites are of great economic importance. The horn

fly, Haematobia irritans (L.), a muscoid fly of Old World

origin, is considered one of the most important

ectoparasites that adversely affects cattle production.

This species arrived from Europe between 1884 and 1886

(Riley, 1889; Marlatt, 1910; Dorsey et al., 1962; Bruce,

1964; Butler, 1990). Both sexes of the fly are obligate,

blood-sucking ectoparasites of cattle but will also feed on

other mammals. Adults feed continuously, taking about 24-38


blood meals a day (Harris et al., 1974). Horn flies spend

their immature stages in the host's manure. Only the adult

stage is parasitic and hematophagous.

Horn fly populations parasitizing cattle in Europe are

relatively low, averaging about 200 flies per animal or

fewer (Hammer, 1942). Extreme numbers, however, have been

observed in the tropics and semitropics, ranging from 1100

per animal for beef cattle to as many as 20,000 for bulls

(McLintock and Depner, 1954). In the USA significant

economic damage to cattle occurs when horn fly density

exceeds 50 flies per animal (Butler, 1975). In contrast,

Schreiber et al. (1987) and Hogsette et al. (1991) reported

the economic injury level to be about 200 flies per animal.

Damage caused by the horn fly includes reduction in normal

weight gain and milk production. Cost to the cattle

industry of horn fly infestation in the USA is about $876

million per year (Kunz et al., 1991).

Since the mid-1940's control of horn flies on cattle

has been based primarily on insecticides. These methods are

effective because the horn fly remains on the host and

leaves the host only to deposit eggs. But the continuous

use of insecticides has resulted in the development of

resistance to many insecticides (Sheppard, 1983, 1984;

Byford et al., 1985; Kunz and Schmidt, 1985; Schmidt et al.,

1985; Cilek and Knapp, 1986; Steelman et al., 1991).

Resistance to insecticides developed rapidly, and efforts


have been made to prolong the efficacy of available

pesticides (Cilek and Knapp, 1991). Because of concerns of

resistance, multiple resistance, cross-resistance to the

major classes of pesticides, ecological awareness and

escalating prices of petroleum-based pesticides, alternative

control measures are receiving greater emphasis, especially

integrated procedures that include anti-arthropod vaccines.

Anti-arthropod vaccine have been developed for tick control.

It works by conferring on the host the capacity to inhibit

parasite attack (Wikel and Allen, 1982; Agbede and Kemp,

1986). Similar work by Schlein and Lewis (1976) on

hematophagous flies shows increased mortality and cuticular

abnormalities when flies feed on rabbits immunized with

crude extracts of fly integuments and other tissues.

In investigating alternative control techniques, it is

essential to understand the interaction between host and

parasite. Information on the immunological aspect of the

interaction could form the basis for the development of an

anti-arthropod vaccine. It may also provide an

understanding of the genetic basis of natural resistance to

parasites. This could lead to the development of livestock

breeds with enhanced genetically based resistance to


Evidence from cattle studies demonstrated that breed

and color influence horn fly infestation, and individual

cows within the same breed show different susceptibilities


to horn flies (Frank et al., 1964; Tugwell et al., 1969;

Holrroyd et al., 1984; Brethour et al., 1987; Cocke et al.,

1989; Steelman et al., 1991 and Steelman et al., 1993). Low

pest populations on individual cows also have been

attributed to dislodging or repelling behavior of hosts as

fly biting density increases (Harris et al., 1987). The

phenomenon of refractoriness or attractiveness of individual

cows within several breeds to ticks has been reported (Riek,

1962; Wilkinson, 1962; Wharton et al., 1969; Seifert, 1971;

Latif, 1984; Latif et al., 1991) and selective breeding has

been used to enhance resistance of host animals to ticks and

lice (Riek, 1962, Clifford et al., 1967, Wharton, 1974,

Sutherst et al., 1979).

Although research has been carried out on the life

cycle and basic biology of the horn flies, little has been

done on the immune response they elicit in hosts. Kerlin

and Allingham (1992) demonstrated that cows exposed to

buffalo flies, Haematobia irritans exigua, developed

antibodies to buffalo fly antigens at levels correlated to

intensity of exposure, but flies fed on animals with high

antibodies in their serum did not show greater mortality

than flies fed on unexposed animals. Development of anti-

arthropod vaccine would require the identification of

specific antigens that can confer immunity.

Investigation of the immunological relationship between

the horn fly and its host and of factors influencing


differential infestations of cows by horn flies could

provide information useful in the development of control

measures that could be used in an integrated scheme.

Control Strategies For The Horn fly

Horn fly attack on cattle causes pain and annoyance

that interferes with feeding and resting. Horn fly feeding

cause laceration, which results in open sores on the head

and under-body of cattle which may lead to secondary

infection caused by pathogens and parasites such as the

screwworm, Cochliomyia hominivorax (Greenberg, 1971). The

piercing-sucking mouthparts (Fig. 1.1) enable the flies to

transmit anaplasmosis and other diseases mechanically within

the herd (Greenberg, 1971).

The importance of methods for horn fly control on

cattle cannot be underestimated in this era of strict

regulation of allowable chemical in food. Control of horn

flies presents unique problems because the immature stages

develop in fresh manure and only the adult stages are

parasitic on the host. The adults' habit of staying on the

host makes it susceptible to chemical control methods. Most

chemical control is directed at the adult or against the

immature stages. Studies show that cattle treated with

insecticides gain weight more rapidly than untreated animals

(Duren and O'Keefe, 1972; Campbell, 1976; Haufe, 1982;

Kinzer et al., 1984 and Quisenberry and Strohbehn, 1984).

Figure 1.1. Scanning electromicrograph of the mouthparts of
the horn fly. Scale bar = 0.2 mm.

Figure 1.2. Scanning electromicrograph of the labium of the
horn fly mouthpart showing the prestomal teeth. Scale bar =
0.2 mm.


Control of horn flies throughout the USA relies heavily

on chemicals. Beginning in the mid 1940s' DDT and other

organochlorines and arsenic chemicals were registered for

horn fly control (Laake, 1946; Matthysse, 1946). Although

resistance to DDT and methoxychlor was later confirmed

(Harris,1964), a variety of chemicals still existed for horn

fly control between 1960 and 1965. In addition to the

organochlorines, organophosphorous compounds such as

fenchlorphosR, crufonateR, malathionR and carbamates like

carbarylR were available. Resistance of flies to toxaphene

and fenchlorphos soon developed. Introduction of

insecticide impregnated eartags initially provided economic

and effective season-long control (Ahrens, 1977; Ahrens and

Cocke, 1979; Sheppard, 1980;). Unfortunately, resistance

soon developed to the organophosphorus insecticide used in

the first ear tags (Sheppard, 1983). StirofosR impregnated

ear tags were then replaced by those containing one of the

pyrethroids, permethrin, or fenvalerate. In the USA,

control problems with the pyrethroids were reported when the

tags had been in use for 2 to 4 years in the southeastern

USA (Quisenberry, 1984). Despite problems, insecticide

impregnated ear tags remain the formulation of choice,

because they fit well in herd management systems and provide

up to 6 months' control of horn flies (Ahrens and Cocke,

1979). The resistance picture of pyrethroid ear tag has

been summarized by Butler (1992) (Fig. 1.3).


Currently, one of the most serious problems facing

agriculture globally is the development of resistance to

most registered chemicals. This often results from the

overuse or misuse of the chemicals. Horn fly resistance is

attributed in part to the elution rate from ear tags, which

diminishes over time (Miller et al., 1984), leading to

exposure of flies to sub-lethal dosages of pesticides. The

loss of pesticide efficacy for horn fly control has made

control difficult and costly, creating problems for both

cattle producers and entomologists because of the limited

supply of replacement pesticides and the tremendous cost for

development of new pesticides (Georghiou, 1986). The cost

of developing new pesticides is now more than the economic

return before development of resistance. Few pesticides are

therefore being developed.

Biological control methods targeted at the immature

stages of the horn fly had some impact on controlling the

horn fly and its sub-species the buffalo fly, Haematobia

irritans exigua (Degeer), from Australia. Worldwide

investigations of natural enemies of the horn fly dealt

with four categories: (1) predators (2) parasites and

parasitoids, (3) pathogens, and (4) competitors.

Introduction of dung-burying scarab beetles from Africa to

augment those in Australia and the United States in the

1970s did not reduce horn fly or buffalo fly populations.

This failure was attributed to the horn fly's greater

Fig. 1.3. Probable distribution of horn fly resistance to
pyrethroid ear tags (Butler, 1992)


ability to disperse to fresh dung compared to the scarabs.

Sufficient horn fly immatures managed to complete

development in dung before scarabs buried them. Climatic

differences between the native African conditions and the

Australian and American conditions also resulted in scarabs

being unable to build up high early season population levels

in their new environments. Burying of the dung also

interfered with the effect of native predators on horn flies

(Macqueen, 1975; Legner, 1978; Roth et al., 1983), and horn

fly populations increased rather than decreased (Harris,


Use of the sterile technique as a component in an

integrated control program was tried for the horn fly in

Brazos County, Texas. Control could not be achieved because

the target area could not be isolated (Kunz et al., 1974).

Eschele et al. (1977) reported the elimination of horn flies

from semi-isolated areas for two weeks by using a

combination of orally administered methroprene insecticide

treatment along with sterile male releases on the island of

Molokai in Hawaii. The feasibility of using the sterile

insect technique for large scale horn fly control is not

known presently.

Possible horn fly control strategies that could be used

in an integrated control program include enhancing

resistance immunologically through immunization, or

developing livestock breeds with genetically based


resistance to arthropods. Much work in this area has been

done with ticks. Trager (1939) was a pioneer researcher in

investigating the possible use of host immune response to

control arthropods. Materials from the salivary glands and

various other internal tissues of ticks were found to be

responsible for inducing tick resistance to animals (Manohar

and Banerjee, 1992). Application of some of the methodology

in the work with ticks could unveil information on horn fly

antigens that might induce acquired immunity in cows. Also,

understanding of the host-parasite relationships could

enable the selection and breeding of cows that are naturally

resistant to horn fly infestation.

Host-Parasite Interaction

Man and animals have been plagued throughout the

centuries by arthropods that consume enormous quantities of

food and are responsible for numerous diseases transmitted

either by them or caused by the insects themselves

(Benjamini and Feingold, 1969). Of these, the blood-feeders

are of great importance because they vector some of the most

dangerous parasites. Little information exists about the

early interaction between parasites and their mammalian

hosts, but it is documented that parasites, together with

their hosts, underwent significant evolutionary radiation.

It is believed that parasitic arthropods developed through a

series of transitional stages from free-living forms to


obligatory parasites. The existing relationships between

parasitic arthropods and mammals are a result of these

interactions during the past thousands of years (Kim, 1985).

Nelson et al. (1975) described ectoparasites as semi-

independent organisms living on host surface but with the

ability to live free from their hosts for short periods or

to move from one host to another. They were divided into 3

groups according to habits or ecological niches: (1) those

feeding only for a limited period on their host and free-

living for most of their life cycle (e.g., Ixodidae:

Acarina); (2) nest ectoparasites most often collected in

the habitat of their host rather than from the host itself

(e.g., Argasidae: Acarina and Siphonaptera) and (3) host

ectoparasites that are permanent residents of the host's

integument (e.g., Anoplura and Mallophaga).

Wakelin (1984) identified nutrient transfer and

energy exchange as central in the host-parasite

relationship. Insect parasites exhibit high levels of

nutritional, physiological and behavioral interactions with

their host, with most ectoparasites feeding on host fluid or

integument. In the host-parasite system, parasites exploit

the host, and, in response, the host reacts to minimize that

exploitation. Blood feeders, like mosquitoes and ticks,

cause the most damage often causing anemia. Blood loss is

correlated with feeding habits of many ectoparasites (Nelson

et al., 1977).

As ectoparasites feed, they also inject their salivary

secretions into the host, triggering host defense responses.

The responses provide new information, and the parasite

make counter-responses. This sequence of responses and

counter responses results in adaptations in behavior,

physiology, development, and even morphology for both the

parasite and host (Whitfield, 1979). Host response to

ectoparasitic attack can be (1) a localized traumatic

reaction to injury caused by the insects' mouthparts, (2) a

toxin injected into the wound in the insect's saliva, or (3)

an immune response to an antigen in the saliva (Lehane,

1991). The response were summarized as immunologic,

allergic, traumatic, toxic or irritant by Nelson et al.

(1977). Physiological consequences of hematophagy on

vertebrate hosts, as reviewed by Nelson (1989), include

metabolic changes, anorexia, inflammation and immune

sensitization, all of which interact to reduce productivity.

Stable fly, Stomoxys calcitrans (L.), and horn fly,

Haematobia irritans attack, cause changes in heart rate and

respiration, nitrogen balance and body temperature of the

host (Schwinghammer et al., 1986a, b).

Lower susceptibility to ectoparasitic, hematophagous

arthropods may be acquired through the host's immune system

(Wikel, 1982). Though hosts' respond immunologically to

arthropod feeding, only in few cases are these responses

associated with development of acquired resistance (Baron


and Nelson, 1985). Host defenses against parasites occur at

both behavioral and physiological levels and the sum of all

the physiological potential of a host that protects it from

parasite infestation is called resistance (Kim, 1985).

Mammalian hosts exhibit both natural and acquired immunity

to arthropod infestation. Levels of acquired resistance

likely reflect a dynamic balance between the immune

rejection mechanisms of the host and the immunosuppressive

capabilities of the blood feeding arthropod (Ramachandra and

Wikel, 1992). Immunity can be developed to parasitic and

predatory arthropods or arthropod fragments (Benjamini and

Feingold, 1969). Acquired immunity to arthropods may be

roughly grouped into three major categories: (1) immunity

associated with the neutralization of toxic substances

introduced into the body by arthropods (eg., venoms of

scorpions, spiders, bees and wasps), (2) hypersensitivity to

antigens of arthropod origin (eg. reactions to arthropod

bites, stings, urticating hairs and spines, allergy induced

by inhaled arthropod fragments, and allergy induced by

ingested arthropods), and (3) immunity to infestation or

invasion by arthropods (eg. infestation by certain ticks and

invasion by certain myiasis-producing dipterous larvae).

This division is of practical value, but arbitrary since

these categories are not necessarily mutually exclusive.

Most arthropods introduce antigens into hosts through

mouthparts. This makes hematophagous arthropods


parasitizing vertebrates the best candidates for

investigation of host immune response (Benjamini and

Feingold, 1969). Apart from disease transmission, the

manner in which hematophagous arthropods obtain their blood

meal often triggers a hypersensitive reaction by the host.

This means that the host responds immunologically to the

arthropod or arthropod products. This reaction may be

unpleasant and could lead to severe manifestations ranging

from pruritus and generalized urticaria to systemic

anaphylaxis and death. The host's reaction to the bite of

the hematophagous arthropod has been studied for many

arthropods, both from the stand-point of the host-parasite

relationship and from the standpoint of its possible role in

disease transmission. By virtue of dependence on the host

for survival, the degree of reactivity of the host to the

bite of the arthropod may play a role in the host-parasite

interrelationship. This may also affect infectivity with

pathogens and their survival in the host. Host

hypersensitivity to salivary antigens of arthropods may

cause severe reactions to bites of a normally mildly biting

arthropod such as the Anopheles mosquito. Trauma caused by

arthropods that lacerate the skin with their blade-like

mandibles (e.g., deer flies, black flies, stable flies, and

tsetse flies) is more severe and noticeable by the host than

that caused by those possessing piercing mouth parts (e.g.,

mosquitoes and fleas) (Benjamini and Feingold, 1969).


Nelson et al. (1977) reported that antigen (oral secretion)

deposited in skin tissue stimulates host's immune system via

the lymphatic route, while those injected intravascularly

probably make their first contact with antibody-forming

cells. He also reported that in many biting flies (e.g.,

mosquito or horn fly) and ticks both the lymphatic and the

circulatory route are involved. Cows exposed to the buffalo

fly (Haematobia irritans exigua) have been documented to

develop serum antibodies to buffalo fly antigens that

correlated with the intensity of exposure (Kerlin and

Allingham, 1992).

Innate host resistance is heritable and includes the

hosts' ability to groom itself, as well as encompassing

inherited factors like skin color and odor. Acquired

resistance describes the ability of the host to interfere

with fly feeding, due to previous exposure or immunization.

The interaction of antibodies with the horn fly is

little understood and the identification of the antigen(s)

that elicit humoral responses in the host cows has not been


Arthropod Defence Mechanism

In nature, ecological stability exists between

parasitic arthropods and their hosts. Insect immune system

developed to protect them from potentially damaging effects

of biological invaders (Lehane, 1991). The ability to


protect themselves by removing microorganisms and parasites

from their circulation has contributed greatly to the

success of insects (Azambuja et al., 1991). Immune

strategies also modulate the effects of host immunologic

effector mechanisms induced by insect parasitism. This

allows the parasite to survive in the face of host

antibodies (Leid et al., 1987). A potent humoral immune

system has been induced by injection of live non-pathogenic

bacteria in insects (Gotz and Boman, 1985). Factors

governing immune reaction are located in the hemolymph in

both cellular and humoral immunity. The insect immune

system is not homologous to vertebrate systems in

distinguishing foreign entities. Using immune reactions of

vertebrates as evaluative criteria would make insects and

other invertebrates immunocompetent (Good and Pappermaster,

1964; Saunders, 1970). Two major features of vertebrate

immune response lacking in the insect immune response are

specificity (antigen-antibody complementarity), and

increased responsiveness to foreign substances as a result

of prior exposure (immunologic memory). Nevertheless,

insects are capable of effectively discerning and combating

foreign antigens (Nappi, 1975). Some insects produce

relatively nonspecific substances (antimicrobial,

antibacterial, lytic) against various invading microbial

organisms that afford the host some defence (Whitcomb et

al., 1974).


Cellular immune mechanisms of insects are mediated by

the cellular components of the insect hemolymph (hemocytes)

that eliminate invading organisms by phagocytosis and/or

encapsulation. Encapsulation defines a coordinated response

involving the aggregation, adhesion, and flattening of

hemocytes over foreign surfaces too large to be engulfed by

individual cells. Generally encapsulation reactions in

insects are accompanied by the intra and extracellular

deposition of melanin pigment (Whitcomb et al., 1974). Most

vertebrate parasites that use insects as biological vectors

are directly or indirectly affected by the hemocoel of the

vectors (Weathersby, 1975).

The immune response in insects is acquired and is an

individual characteristic. It is of little practical use to

control vectors. Many vertebrate parasites that might be

expected to produce acquired immunity in insects either do

not penetrate to the hemocoel or they cause very high

mortality in the vectors. Often, the vector only lives long

enough to transmit the causal agents of diseases and there

is little opportunity to determine immunity. Some insects

may acquire the ability to transmit a parasite to offspring

but it is not likely that an antigenic stimulus will affect

the organism genetically. Although insects produce specific

substances in response to foreign substances and antigenic

stimulation, they appear to have little resistance to

overcoming many parasites (Huff, 1940).


The description of the arthropod immune system as

primitive and unsophisticated, compared with that of

vertebrates, is becoming untenable. At least six

monosaccharides or their derivatives found on the plasma

membrane of vertebrate cells are present on the immunocyte

plasma membrane of one of the most primitive arthropods, the

horseshoe crab Limulus polyphemus (Gupta, 1991). They most

likely participate in foreign tissue recognition,

phagocytosis, and encapsulation, and at least one (sialic

acid or NANA) is involved in signaling and cell-cell

recognition and activation of the complement pathway

(Gupta, 1991). Furthermore, specific inhibitors of the

classical and alternative complement (C) pathway of

mammalian complement are present in the insect hemolymph.

It is now thought that the key component of the vertebrate

complement exists in insects. Evidence is accumulating that

in some insects, a humoral adaptive response that possesses

specificity and memory comparable to the functional

attributes of the mammalian immunoglobulins exists.

Extensive studies have been done on humoral immunity in

lepidopterans, but relatively little has been done on

dipterans (Kaaya et al., 1987).

Vertebrate Immune System And Host-Immune Response.

The immune system of vertebrates includes bone marrow,

thymus, bursa of fabricius, spleen, lymph nodes, gut-


associated lymphoid (GALT), and the reticuloendothelial

system (Clark, 1986). This system performs the task of

protecting the host from threats to homeostasis, both

exogenous and endogenous (Bigley et al., 1981). A

coordinated system of cellular communications enables host

cells to differentiate between "self" and "non-self", and

react appropriately to eliminate foreign antigens (Nappi,

1975). Depending on the nature of the antigen, one of two

major forms of immunologic effector mechanisms may

predominate in destroying or eliminating the antigenic

material. A substance which when introduced into an

organism induces an immune response is called an antigen.

Its immunogenicity refers to its capacity to stimulate an

immune response under a set of conditions. Severity and

type of antibody response in hosts depends on antigen

chemical structure, physical state, stability, size,

frequency of presentation, route of presentation (oral,

dermal, respiratory etc.), complexity of the array of

immunogens, and presence of other factors (e.g presence of

adjuvant that allow antigens to remain at site) and

idiosyncrasies of the host (Spiegelberg, 1974). Response to

potential antigens varies from no response e.g. high-zone or

low-zone tolerance (Clark, 1986) to severe allergic

hypersensitive reactions. Once detected, invading antigens

are identified, and processes for their elimination are



The vertebrate immune response is made of two systems:

the humoral and cellular systems (Clark, 1986). In the

humoral system a specific B-cell recognizes and interacts

with an immunogen, it is then activated to differentiate

into a plasma cell which then undergoes multiplication and

forms clones of the original plasma cell. The resulting

plasma cells synthesize and secrete specific glycoprotein

molecules (the antibody or immunoglobin) into the blood

plasma. Some antigens stimulate the thymus gland, and this

results in the appearance of "T" lymphocytes, with effector

activities such as the ability to kill cells by membrane

contact or to elaborate soluble products that assist in the

development of inflammatory responses. In the cellular

response, a "T" cell recognizes and binds to the antigen

leading to the ultimate elimination of the antigen from the

system (Clark, 1986). Mammals also utilize immediate and

delayed mechanisms for protection (Raven and Johnson, 1987).

The immediate mechanism functions by first isolating the

entrance site through the inflammation process (Wakelin,

1984). The immune response then attempts to destroy the

invading organism by phagocytosis, utilizing macrophages and

granulocytes such as neutrophils, eosinophils, and

basophils. The delayed system, consisting of a humoral or

B-cell system and a cell mediated or T-cell system, is

turned on by the above phagocytosis, forming antibodies and

sensitized lymphocytes.


Host response to arthropod bites varies among species,

among individuals of one host species, and also with

arthropods doing the biting (Nelson et al., 1977). A

general sequence of skin reactivity has been noted in time

in hosts repeatedly exposed to insect bite and has been

broadly divided into five phases. These phases of the

sequence were described from studies of rabbit and human

responses to mosquito bites (Mellanby, 1946; McKiel and

West, 1961) and to guinea pig responses to flea bites

(Benjamini et al., 1960). Feingold et al. (1968) described

the phases as follows:

Phase I. Induction. No response or reaction, but foreign

material is identified by Langerhan's cells and macrophages

as antigenic.

Phase II. Delayed hypersensitivity reaction. Dermal

responses to antigen appear 24-48 h after exposure and

include redness, swelling, and itching at site of exposure

(Dahl, 1987). Delay is caused by time required for specific

leukocytes to migrate to site of exposure and release

lymphokines that recruit other leukocytes and cause

capillaries to vasodilate, resulting in edema and

infiltration of erythrocytes.

Phase III. Delayed and Immediate hypersensitivity reaction.

Delayed and immediate reactions occur jointly, with

immediate reactions causing redness and swelling within 15

minutes (Clark, 1983; Dahl, 1987). Immediate responses are


rapid because plasma cells elaborating immunoglobulin class

E (IgE) are present. The IgE binds to Fc receptors of mast

cells in the dermis and are available for immediate binding

with antigen.

Phase IV. Immediate hypersensitivity reaction. Delayed

responses have waned, perhaps because of deletion of T-

helper cells (Dahl, 1987), or more likely because T-

suppressor cells have appeared and are suppressing the

delayed response (Dahl, 1987, 1989; Van Neste, 1988).

Phase V. Desensitization or Hyposensitization or no

reaction. No hypersensitivity occurs on exposure to antigen

perhaps because of the effect of T-suppressor cells (Dahl,

1987, 1989) or because IgG is blocking the IgE-mediated

reaction by direct competition for antigen (Dahl, 1989).

Here, antibody-producing cells have been exhausted or are

suppressed. These groupings are not mutually exclusive.

The development of acquired host immunity to arthropods

as a means of ectoparasite control is of scientific interest

(Preutt and Thomas, 1985). Research by Wikel (1982)

suggests that hosts' lower susceptibility to ectoparasitic,

hematophagous arthropods may be acquired through the host

immune system. Results of immunoglobulin participation in

the development of acquired immunity could be associated

with host skin reactions that disrupt feeding efficiency or

react with target antigens within the gut or hemocoel of the

arthropod interfering with digestion or metabolism. Primary


targets might include salivary gland secretions, intestinal

symbionts, digestive enzymes or secretions of the gut such

as digestive enzymes or the peritrophic membrane (Pruett and

Thomas, 1985).

Host Location By Arthropods

The survival of parasitic species depends on their

selection or location of suitable hosts. Host location is a

difficult and complex behavioral task involving an

integrated but flexible behavior package which gathers

momentum as a host is tracked down (Lehane, 1991). Species

that actively search for hosts move towards stimuli, while

passive species wait either on vegetation or within refuge

sites until a host comes within reach (Waladde and Rice,

1982). The sequence of behaviors involved in host searching

are susceptible to manipulation or interference by humans

(Colvin and Gibson, 1992). These behaviors though do not

occur in a strict and inflexible sequence (Lehane, 1991).

The effectiveness of chemical control of adult flies is

based on the knowledge of their host seeking and resting

behavior. Generally only exogenous factors can be

manipulated in field populations. Host seeking and location

in insects is guided by both chemical and physical cues

originating from the environment and is elicited by visual,

thermal, humidity, and chemical stimuli operating separately

or in combination (Laarman, 1955; Dethier, 1957; Brown,


1966; Gillies and Wilkes, 1969; Hocking, 1971; Vale, 1977;

Dalton et al., 1978; Kinzer et al., 1978). Some nonchemical

factors influencing host selection and location include

wavelength and intensity of light reflected from the host.

Two major sensory inputs of recognized importance are

olfaction and vision. The importance of vision in host

location has been demonstrated (Parr, 1962; Brady, 1972;

Vale, 1974). The response of blood-sucking flies to carbon

dioxide and host odors has been studied in the field (Vale,

1980) and in the laboratory (Gatehouse and Lewis, 1973;

Mayer and James, 1969). Host odor (Hargrove and Vale, 1978)

and the odors of excretory products (Owaga, 1984; 1985) have

been found to be highly attractive to tsetse (Glossina

spp.). Identification of components of such odors has led

to the development of effective baits for sampling or

controlling tsetse populations (Vale et al., 1986). The

chemical, l-Octen-3-ol, was isolated from cattle odors and

has been found to be a potent olfactory stimulant and

attractant for tsetse (Hall et al., 1984). Krijgman (1930)

conducted experiments with the stable fly, Stomoxys

calcitrans, in a simple olfactometer in still air and

reported orientation to the odor of fresh horse blood. The

same species failed to respond to olfactory stimuli from

blood when the odor was dispersed by moving air (Gatehouse

and Lewis, 1973). Tests of various components and fractions

of blood as attractants resulted in the discovery of an


extremely volatile constituent which attracts Culex

mosquitoes and Stomoxys. It appears that the material acts

as an attractant to which the insects orient, and as a

release for the act of piercing. It is believed that this

volatile fraction of blood diffuses through the skin of the

host and is an important factor in attracting mosquitoes and

biting flies to the host (Galun, 1975a). Attractants can be

either close range or distant depending upon how far away

stimulation of olfactory response occurs from the source.

It is uncertain how far odor signals travel before they fail

to be detected. In nature, horn flies show preference for

attacking some cow breeds and color, whereas, others are not

attacked even though they are close to those attacked. Upon

eclosion, horn flies depend more on vision than olfactory or

host stimuli to locate their hosts (Hargett, 1962; Hargett

and Goulding, 1962; Milstrey, 1983). Horn flies are

influenced by light stimuli and negative geotaxis, and fly

up into the air. Kinzer et al. (1978) observed that

temperature and Co2 were prime factors in horn fly

orientation. Dalton et al. (1978) found radiated rather

than internal heat and host color to be the most important

factors influencing flies, especially in close quarters.

Milstrey (1983) observed that horn fly population and

migration was regulated in part by semiochemicals. Bolton

(1980) and Mackley et al. (1981) showed that in horn flies 4

olefins, (Z)-9-tricocene (masculure), (Z)-5-tricocene, (Z)-


9-pentacocene and (Z)-9-heptacocene serve as mating and

aggregation pheromones. Milstrey (1983) showed that blends

composed of equal parts of all 4 olefins and a combination

of (Z)-9-heptacosene had little effect. An olefin blend

(50:40:6:4) of (Z)-tricocene, (Z)-5-tricocene, (Z)-9-

pentacocene and (Z)-9-heptacocene best reflected natural

pheromone levels in the field. This blend significantly

reduced horn fly numbers on cattle and attracted them on

horses for a short period of time. Horn flies are thought

to lay pheromone trails which in higher concentrations

causes flies to migrate. Chamberlain (1981) showed that

horn flies locate their host by attraction not by passive


Hosts possess characteristics which influences parasite

infestation, and these have been exploited for some pest

control. Detection of odors by insects is presumed to

result from changes in the electrical activity of primary

olfactory receptor neurones contained within the antennal

sensilla (Grant and O'Connell, 1986). The majority of

studies on odor-mediated flight orientation in insects has

involved upward flight of male moths toward a source of

female pheromones as a model (Kennedy, 1986). Almost all of

these observations have been made in laboratory wind

tunnels. Attractive compounds that have been identified in

cattle odor for tsetse flies include carbon dioxide, acetone

and l-octen-3-ol (Bursell, 1984; Hall et al., 1984; Vale


and Hall, 1985). The screening of compounds for their

olfactory effectiveness by fly trapping in the field is time

consuming and subject to many variable factors. Olfactory

components are responsible for these effects. The

components can be exploited for control purposes.

The complex behavioral repertoire culminating in host-

location by hematophagous arthropods involves an array of

chemical and physical cues (Chapman, 1961; Hocking, 1971;

Friend and Smith, 1977). It is now well documented that

carbon dioxide emanating from the host constitutes a primary

chemoattractant to blood sucking ectoparasites (Garcia,

1962, 1965; Wilson et al., 1972).

Role Of Saliva In Blood-Feeding By Arthropods

In their quest for a blood meal, hematophagous

arthropods introduce an array of salivary compounds into the

host. The saliva stimulates host immune response alerting

the host to the presence of an arthropod. The immune

response ultimately may deny a blood meal to the arthropod.

Salivation may look detrimental to an insect, but studies

have identified some benefits to the insect (Lehane, 1991).

The function of saliva in many important insect vectors are

poorly understood (Kerlin and Hughes, 1992). These authors

reported that probable salivary secretions from parasites

promote feeding and survival on the host. Many functions

have been attributed to the saliva of blood-feeding


arthropods. The first array of digestive enzymes is found

in the salivary glands. Among other things, the contents of

the saliva facilitate blood feeding by blocking the

hemostatic reactions or by causing allergy hemodynamics.

Saliva from Ixodes damini have been found to contains anti-

hemostatic, immunosuppressive, anti-inflammatory, and

neutrophil inhibiting components (Ribeiro et al., 1985).

Anticoagulants which prevents coagulation of blood which is

capable of blocking the mouthparts of the insect have been

found in saliva or salivary gland homogenates of certain

hematophagous arthropods but not in others (Gooding, 1972).

Also hemagglutinins have been found in the salivary glands

of some hematophagous arthropods but their roles remain

unknown (Gooding, 1972).

Most pathogens transmitted by ticks are introduced into

host bodies with saliva (Hoogstraal, 1970; Binnington and

Kemp, 1980). Tick salivary gland secretary products

include: (1) cement to help anchor the mouth parts, (2)

anticoagulants in some species, and (3) antiinflammatory and

immunosuppressive molecules that assist in feeding and

evading host defense mechanisms (Binnington and Kemp, 1980;

Ribeiro et al., 1985). Host immunity can be elicited in

response to antigens secreted by the salivary glands. The

size, mass, and protein content of the salivary gland

increase approximately 25-fold during tick feeding (McSwain

et al., 1982).


It has been recognized that for any blood-feeding

ectoparasite, the range of antigens presented to the host in

the normal feeding process is likely to be very limited.

The possibility exists of inducing specific immune

protection in the host with parasite molecules other than

those presented in normal feeding (Galun, 1975b; Ackerman et

al., 1980; Mongi et al., 1986; Willadsen, 1987). The range

of potential targets for such artificially primed

immunological attack on the parasite by the host is very

wide and may be specifically targeted towards disrupting

selected tissues or physiological processes. (Essuman et

al., 1992).

Immunochemistry And Antibodies As Investigative Tools

Immunohistochemistry techniques are valuable tools

employed to detect antigen and antigen sources through the

use of specific antibodies that are labelled so that the

sites of antibody attachment becomes microscopically visible

and still preserve anatomical details. Fluorescent antibody

techniques first introduced by Coons et al. (1941) have been

used widely on fresh or frozen tissue specimens until

recently. These produce labile staining that can be

visualized only with an ultraviolet microscope. Enzyme-

labelled antibodies and methods applicable to tissues fixed

in standard fixtures such as formalin have been more

recently being employed in immunohistochemistry. These


techniques create permanent stains that shows the

distribution of antigen-antibody complexes under ordinary

light microscope (Haines and Chelack, 1991). The

underlining principle of these techniques is that labelled

antibodies react with antigens without interference with the

biological or immunological properties of the proteins.

Figure 1.4 shows commonly used immunoenzyme staining

methods. Methods in which the primary antibodies specific

for the antigen of interest are labelled with an enzyme are

termed direct methods (l.4a). In these methods, the

antiserum is incubated on the tissue followed by addition of

an enzyme substrate that causes the deposition of an

insoluble colored reaction product at the sites of antibody

binding in the tissues. This reaction product is visible

with light microscopy. Direct immunostains are simple and

economical to perform; however, they provide little

amplification of the visible signal, so they are useful when

antigen levels are high. An additional disadvantage is that

each primary antiserum must be conjugated to an enzyme.

Indirect immunostain methods utilize an enzyme-

conjugated anti-immunoglobulin second antibody to detect

binding of the primary antibody to the tissue section (Fig.

1.4b). Although indirect immunostains are somewhat more

complex and time consuming to perform, these stains have two

advantages over direct methods. Firstly indirect stains

enhance the sensitivity of antigen detection because several


secondary antibodies will bind to each primary antibody.

This intensifies the visible signal produced by the binding

of each primary antibody. Secondly, indirect stains do not

require conjugation of each of the primary antisera.

In addition to the indirect methods there are a variety

of other immunoenzyme techniques designed for greater

amplification of the visible signal produced by the binding

of primary antibodies to tissue sections. One of the most

versatile of these techniques is the avidin-biotin complex

(ABC) method (Fig.l.4c). The ABC immunostain relies upon

the high avidity of the B-group vitamin, biotin, for the egg

white glycoprotein avidin. Antigens in tissue sections are

incubated with an unlabeled primary antibody followed by a

second antibody conjugated with biotin. Following exposure

to the second antibody, the tissue is subsequently reacted

with avidin enzyme substrate complex. Each avidin molecule

has binding site for 4 biotin molecules. The avidin-biotin

complexes are produced, ensuring free biotin-binding sites

on the avidin molecule, which promotes binding of the

complexes to tissue-associated biotinylated second antibody.

The antibodies are labelled with an enzyme, usually

perioxidase. As in direct and indirect immunoenzyme

methods, an enzyme substrate is applied to the tissue and a

colored reaction product forms on the slide at sites of

antibody-enzyme complex binding. The ABC method is

technically complex, time consuming and expensive to perform

compared to direct and indirect techniques. However the

amplification afforded by this procedure is often necessary

to detect antigens in low concentrations or antigens

immunogenically altered by formalin fixation.

Analytic immunologic procedures have become important

parts of the arsenal of techniques for describing and

elucidating physiologic and developmental changes in natural

occurring antigens from insects. Antisera, containing

specific antibodies have been used as analytic reagents to

make qualitative as well as quantitative measurements of

these insect antigens. The presence of common or cross-

reactive antigens can make specific serological detection or

diagnosis of animal exposure to a given agent a difficult

task. The development of several immunological and

serological techniques have been of great help in the

identification and characterization of specific antigens

from complex mixtures.

The source of antigen can be located or localized by

immunological probing with antibodies. Both monoclonal and

polyclonal antibodies can be used for probing. Various

immunological techniques have been developed for probing

(Kobayashi et al., 1988). The advantage of immunologic

detection of antigen is that it does not necessarily require

isolation and culture of specific antigens or tissues

producing them.

Monoclonal Antibodies

Monoclonal antibodies are homogeneous antibodies

derived from a single clone of hybrid cells. They recognize

only one epitopes of an antigen (Kennett, 1979; Goding,

1980; Edwards, 1981). Kohler and Milstein (1975) first

developed antibody producing hybridoma cells. Antibodies

produced in mice, and lymphocytes from the spleen, source of

the antibodies, are fused with mouse myeloma cells in

culture. The fusion allows the hybrid cells to continue to

grow and divide in culture and also to produce antibodies.

One hybridoma cell produces one specific antibody. These

specific antibody producing cells are selected and cloned.

ELISA (Enzyme-Linked Immunosorbent Assay)

The principle behind the ELISA test is the detection of

an antigen with enzyme labelled antibodies (Ma et al.,

1988). The enzyme moiety keys a colormetric reaction that

is proportional to the amount of antigen molecules bound to

the enzyme-antibody conjugate.

There are three ELISA protocols that are frequently

employed for the analytic measurement of a chemical in

biological samples. They are (1) competitive ELISA, (2)

direct double-antibody sandwich ELISA, and (3) indirect

antibody sandwich ELISA (Voller et al., 1979).

In the competitive ELISA procedure, enzyme-labelled

antigen is used to compete with the nonlabelled antigen

present in a sample preparation for available antibody

sites. This protocol is effective if the antigen is stable

and can be easily purified in milligram quantities. The

amount of bound enzyme-labelled antigen can then be

estimated after the addition of enzyme substrate and

subsequent colorimetric reading.

The double antibody sandwich ELISA uses an enzyme-

labelled antibody that is also specific to the antigen. An

antigen molecule is "sandwiched" between the coating or

primary antibody and enzyme-labelled secondary antibody.

The indirect double-antibody sandwich ELISA differs

from the direct method in that the antigen is sandwiched

between specific antibodies developed in two different

animal species. If the coating antibody is developed from

guinea pig, then the secondary antibody must be from another

species such as rabbit. The amount of secondary antibody

bound to the antigen is, in turn, detected by a commercially

available enzyme-AB conjugate specific for the

immunoglobulin of the secondary antibody; in this example

amount of bound enzyme-labelled antigen can then be

enzyme labelled goat anti-rabbit will act as the tertiary


Research Objectives:

1. To investigate the attractancy or repellency of hexane

washes of cows deemed refractory and susceptible to horn


flies infestation using a ten-port olfactometer.

2. To study the host antibody response to horn fly salivary

proteins in naturally exposed mice and immunized rabbits.

3. To identify and characterize immunogenic secretary

proteins from horn fly salivary glands.

4. To utilize monospecific polyclonal antibodies produced

in rabbits and mice to identify and localilize sites of

secretary antigen production in situ utilizing

immunohistochemical techniques.


enzyme substrate


enzyme substrate

enzyme conjugated
primary antibody

_/ \. antigen
*-- tissue section
glass slide

enzyme conjugated
secondary antibody
Primary antibody
S g tissue section
b glass slide


glass slide

Figure 1.4. Diagrammatic representation of three enzyme
immunohistochemical staining methods, a. Direct immunostain.
b. Indirect immunoenzyme stain, c. Avidin-Biotin complex
(ABC) stain (Haines and Chelack, 1991).



Blood-feeding insects locate and choose their hosts by

responding to a variety of olfactory, visual and thermal

stimuli (Vale, 1977; Dalton et al., 1978; Kinzer et al.,

1978), but there is evidence that vision is most important

(Hargett, 1962; Harget and Goulding, 1962; Milstrey, 1983).

About an hour after eclosion, adult horn flies begin to

search for a host. Heavy infestations have been observed on

dark or black cattle compared to lighter colored cattle

(Bruce, 1964). Some animals even of the same hair color,

attract more horn flies than other animals (Schreiber and

Campbell, 1986; Steelman et al., 1993). Bulls attract more

flies than steers or cows (Dobson et al., 1970).

Understanding the factors that influence differential

infestation of cows could lead to the development of more

effective control measures. Designing laboratory

experiments to demonstrate attractiveness or repellency to

different odors is difficult, primarily because horn flies

exhibit strong phototaxis (Morgan, 1966; Kinzer et al.,



1970), making it difficult to balance light variables in a

conventional "Y" tube olfactometer. A "Y" tube olfactometer

was successfully used for horn flies by Bolton (1980).

One of the problems encountered in arthropod olfaction

studies is the difficulty in measuring and quantifying

arthropod behavior to different odors. The "Y" tube

olfactometer, which has been used for most olfaction

studies, offers flies only two choices (Bolton, 1980). This

limits its use in evaluating the response of flies to

multiple odors at the same time. Refining or replacing the

present methods of evaluating chemicals would be difficult

and expensive. Recent technological advancement in

electronics and computers, especially in large-scale

integration, has resulted in powerful microcomputers at

affordable prices that are comparable with, or less than,

conventional activity recording equipment (Symonds and

Unwin, 1982). Advantages of these systems have helped in

the design of an olfactometer capable of testing arthropod

response to multiple chemical stimuli without human

interference by Dr. J. F. Butler of the Department of

Entomology, at the University of Florida (Fig. 2.1) with

support from International Flavors and Fragrances, Inc.

(IFF, NJ). The patented olfactometer (US Patent: 4,759,228;

5,134,892; 5,118,711; 5,165,026; 5,175,175) is

electronically monitored and functions as an activity,

feeding, and touch detector by offering individual horn


flies the choice among 10 different odors presented

simultaneously on ten artificial hosts. The odors are

carried in an airstream moving at 375-500 cm/min. The

olfactometer creates distinct and contagious odor fields

that can be easily entered, left and reentered by the horn

flies seeking a food source. The indices of horn fly

response to the odors are recorded on a computer as logged

touch and bite contact seconds in a time series of up to 4


The purpose of this study was to evaluate the response

of horn flies to hexane washes of repellent and attractive

cows from Arkansas (supplied by Dr. C. D. Steelman through

S242 program) and to evaluate the effectiveness of the

olfactometer to screen natural chemicals from IFF for future

field assays.

Materials And Methods

Individual cows from Arkansas were classified as either

resistant (refractory) or susceptible to horn flies by Dr.

Steelman, University of Arkansas, based on fly count

records. Cows were classified by recording fly populations

on individual cows during the summer season for multiple

seasons. Hexane extracts of cows were collected during the

winter and spring seasons when summer fly pheromone residues

would have lost their lasting effect. This period

avoids the influence of fly factors which occurs when flies

Fig. 2.1. Olfactometer for laboratory assay of hexane
washes of attractant and repellent cows against horn flies
(UF/IFF Patent No 4,759,228)


are in large populations. A wash of each animal was made by

pouring 500 ml of spectograde hexane on the shoulder and

side of animals. The rinse was collected in a metal paint

roller pan and concentrated at room temperature under NO2

until the volume was 2-3 ml. The concentrates were put in

glass screw cup vials (1.5 cc diameter x 6.5 cm high),

covered with aluminum foil, labelled and shipped overnight

to the University of Florida where they were stored in an

explosion proof refrigerator at 4C.

Pretests were conducted to identify those animal

extracts exhibiting extreme refractoriness or attractiveness

to laboratory-reared horn flies. Eight hexane washes were

randomly selected from cows classified as refractory or

resistant. Another eight were selected from those

classified as susceptible or attractive. A comparison was

then made between 10 samples made up of 8 cow washes and 2

standards. An attractant standard (IFF.3145) and repellent

standard (IFF.3380) supplied by International Flavors and

Fragrances (521 West 57 Street, New York, N.Y. 10019).

Attractant and refractory cow washes were compared

utilizing the UF/IFF patented pie type olfactometer. Two

sets of trials: (1) for the refractory cows and (2) for the

susceptible cows were run to select the extreme attractant

and extreme refractory cow washes. The eight selected

washes from each group along with IFF.3145 and IFF.3380 were

randomized within each replication and assigned to the 10


ports of the olfactometer. Each trial was replicated three

times. Analysis of variance (ANOVA) statistical analysis

(Excel package) was used to select the most active of the

susceptible washes and the least attractive in the

refractory washes. Four of the most attractive washes and 4

of the least active washes were randomized with the standard

attractant (IFF.3145) and standard repellent (IFF.3380) and

replicated four times in the olfactometer. After running

the washes eight, IFF chemicals identified as repellents

were run with the standard attractant (IFF.3145) and

untreated air (1000) against horn flies to test the

efficiency of the olfactometer in assaying chemicals. Five

replicates were run and ANOVA was run on data. Statistics

comparison were utilized to determine differences between

the standards and the animal extracts at the 4 hour trial

period. All trials were run for 4 hours.

The olfactometer chamber acts as a slight attractant

for insect movement to the perimeter of the choice chamber.

Stimuli other than olfaction, such as light, air flow,

temperature of the air and surface, humidity, and detector

position, were standardized to aid insect movement to

treatment choice sites (UF/IFF pat. No 4,459,228). Because

of the ultrasensitivity of horn flies to light

differentials, the tests were conducted in total darkness.

The olfactometer uses a center point for air exhaust and

side ports for presentation of treated air. The odor


sources were presented separately through tygonR (6 mm

inside diameter X 17 cm long) tubing attached to the

olfactometer and agar block (Fig. 2.3). Agar blocks,

sensors, tygon and plate surfaces were replaced after each

trial to avoid contamination. Surfaces exposed but not

replaced were washed with hot water and sparkleen

biodegradable detergent.

Processed air and treatments were introduced through

the perimeter ring at 10 ports. Five (5) ul of the hexane

washes were pipetted onto a polystyrene disc containing agar

covered with a silicone membrane (Butler et al., 1984) and

onto polyethylene interflow pellets (Cromex Corp. NY) which

were placed within a tygon tube. About 60 to 100 general

(24-48 hour old) flies of mixed sexes were introduced into

the olfactometer after having been immobilized by cooling at

10C for about 5 minutes. Each trial was run for 4 hours.

Fly preference and contact activity were monitored by

datalogging the time in bite/contact seconds the flies made

contact with sensors (Fig. 2.2), which was then recorded in

bits per minute by a computer. Arthropod contact in the

olfactometer was monitored and detected by a 10 channel

system through Stawberry treeR programming using workbench

3.1 Mac.II (This was specifically adapted to the

olfactometer by S.A. Butler, Gainesville, Florida).

Strawberry tree program was set up to sense and record

feeding through a custom made amplifier (J. Greenberg,


Gainesville, Florida) operating at 100X which detected

differential signals produced by the insect feeding through

the artificial host (agar block) membrane system (Fig. 2.3).

Data were analyzed by logging data into an MS excel

spreadsheet utilizing a macro which calculated the total

number of seconds flies made biting contact during a given

time period (bites second) at 10 or 60 minutes intervals on

each channel. The experiments were conducted in a roomsize

Faraday cage (Lindgren enclosures, Model No 18-3\5-1) in the

dark (Fig. 2.4). Cow washes and chemicals are listed below.













Refractory or resistant cows:





Susceptible Cows:





Chemicals from IFF

620 IFF.3085

381 IFF.3022

219 IFF.2621

218 IFF.2251


Figure 2.2. Horn flies attracted to odor or chemical at a
port sitting on sensor.

'.tnt ;:g~;

ftftW!P.- ^ -di

Figure 2.3. Hook-up of treated artifial host (Agar packet).

Figure 2.4. Patented olfactometer in operation in the
Faraday cage under fiberoptic light conditions.


Figure 2.5 shows the comparison of the IFF standard

attractant (3145) and standard repellent (3380) and the

gradation of fly activities during the 4 hour runs for the

different refractory cow extracts. There was a significant

difference in treatment and interaction. Figure 2.6 shows

contact second rates for the attractive animal extracts.

There was significant differences between replication,

interaction and treatment at the P,0.05. Refractory cow

washes tested, with activity close to that of the IFF

standard repellent (IFF.3380) (Patent applied for) were

considered as strong repellents and selected for further

assay. Hexane washes from the susceptible cows tested, with

activity similar to that of the known standard attractant

(IFF.3145) were considered to be strong attractants and

selected for further assay. By ranking, 2876, 33, 1711 and

48 were considered strong repellents and 17a, 28, H15 and

3020 were considered strong attractants.

Results of the trial runs of the combination of the 4

selected strong susceptible washes and the 4 selected strong

refractory washes with IFF.3380 and IFF.3145 are shown in

Fig 2.7. Hexane wash 2876 maintained its refractory nature.

Hexane wash H15 maintained its susceptible or attractant

nature. Hexane washes 33 and 1711 reversed their earlier

refractory nature and became attractants. Hexane washes 28,

3020 also reversed susceptible nature and became strong

repellents. Washes 48 and 17a which were refractory and

attractants respectively, exhibited, near neutral behavior.

Figure 2.8 shows the results from the trial of repellent

chemicals supplied by IFF. The IFF.3145 was significantly

different from all the materials tested. But there was no

significant difference between the chemicals tested. The

interaction, replication and treatments were also

significant. The chemicals and the air (1000) acted as

repellents as described by the suppliers IFF.


The results of the tests confirms the effectiveness of

the multi-port olfactometer as an assaying tool. The

results also show that there is a basis for labelling

individual cows as susceptible (H15) or refractory (2876) to

horn flies. From the tests conducted it is likely that the

responsible factors are located in the skin of the cows, but

additional tests are required to determine the specific

factors involved. Discrepancies occurred with some

chemicals reversing roles, especially with the trial

combining both attractants and repellents. This may be

attributed to the "nearest neighbor effect" and interaction.

This means that a stronger attractant may be able to pull

flies away from relatively weaker attractants resulting in

the relatively weaker attractants acting more neutral or

refractory. It may also be attributed to the active

Source of. VariationF Pvae Fct.05






8.50641 E-08


03380 B2876 33 E]1711 B 17

048 1 3607 271 0 104 M3145

3160 48417117:1 200 Time
17424 17 1711 33 2 61:00 Series

Skin & Air Treatment @ 51l 2876 3380

Figure 2.5. Activity of horn flies exposed to hexane washes
of resistance or refractory cows. Numbers refer to hexane
washes of cows.

F P-value F criL 0.05

Source of Variation

Source of Variation SS




df MS F P-value Fcrlt 0.05
2 5097416.904 22.43507156 5.30862E-08 3.150411487
9 3100435.713 13.64583247 1.31044E-11 2.040096092
18 1003326.34 4.41590293 6.83283E-06 1.778445835
60 227207.517


3145 17a I '!P 1B a
28 H15 3020 2876a 36

Skin S Air Treatment @ 51d

34 138 3380

a 2876a
M H15
B 28
E3 17a
0O 3145

00 -rune
1:00 Senes

Figure 2.6. Activity of horn flies exposed to hexane washes
of cows identified as susceptible or attractive to horn
flies. Numbers refer to hexane washes of cows.

Source 01 VariatIon F P-value F criL 0.05






3.27831 E-25


D13380 028 *2876 B3020 M17a B48 I H15 1711 *233 8 3145

1(11 1H15 48 ^
H15 48 17a
Skin &;'Air Treatment @ 5;1


3: 4:00
5 3:00 -
02:00 Time
1:00 Series

Figure 2.7. Activity of horn flies in the presence of
selected refractory cow and susceptible cow washes.
Numbers refer to hexane washes of cows.

Source of Variation

F P-value F criL 0.05

3620 3381 3219
32 T n 3213.31 42255 9

Skin TreaLmenl 0 0.005g 2256 1000

Figure 2.8. Results of testing repellents. Numbers refer
to chemicals supplied by IFF.

13 22o

D* I ^-'vo


1 * 400
ZA^ 3 00
.00 0
: 100

ingredients in some of the washes breaking down and changing

the characteristic of the compound.

In the olfactometer runs, it is possible for different

plumes from chemicals or odors offered to the horn flies at

different ports to overlap and interact, resulting in the

flies making ill-defined chemotactic responses. To avoid

this or minimize its effect untreated air was piped between

each treatment to keep plumes separate. The untreated air

stream also acted as a valve for insects exposed to higher

than acceptable odors or chemicals thus acting to reduce

attenuation of sensors. This might not occur in a field

experiments since the insects will have the option of moving

away from odors that are objectionable. To analyze the data

from the runs, the total of attractiveness or repellency was

logged as a mean over a 4 hour time period. Time of non-

touching or non-feeding gaps in the feeding pattern were

averaged out with the periods of feeding on bite second

intervals at 10 minute or 1 hour blocks.

Testing of the 8 semiochemicals samples from IFF show

that they were all highly significant repellents as compared

to the attractants. This confirmed the reliability of the

olfactometer to screen multiple chemicals for further tests,

and helps in avoiding costly and morally challenging use of

animal or human subjects. Air (1000) served as a neutral or

non-reacting treatment and offered an escape for the flies

when overwhelmed or attenuated with repellent chemicals.


Some chemicals are known to maintain a designated fly

response description during the initial stages of a trial,

then change to an opposite designation with time, this may

be ascribed to dosage shifts as materials change or are

eliminated or behavioral shifts in the flies. With this

observation it will be wise to thoroughly evaluate an

attractant or repellent under actual field use before

recommending widespread use.

A wide variety of host signals are involved in host

finding (Lehane, 1991). The natural refractiveness or

attractancy of individual animals to pest infestation has

been recognized and has been used to develop livestock lines

or breeds with enhanced, genetically based tick and lice

resistance (Reik, 1962; Clifford et al., 1967; Wharton,

1974; Sutherst et al., 1979). Horn fly differential

infestation of cows, as observed by Steelman et al. (1993)

and confirmed by the olfactometer represents an example of

natural host resistance to ectoparasitic infestation which

need further investigation to determine the factors


Control of an arthropod can be effective only by

knowledge of the biology of the target species, in this case

the factors used by the horn flies to locate and parasitize

their host. Visual and olfactory stimuli, aided by

anemotactic and optomotor responses, are considered the most

important signals when the insect is still at some distance


from the host. Nearer to the host, different stimuli become

important, particularly heat (Lehane, 1991). Despite the

importance of individual stimuli, multiple stimuli are

likely to be a better guide to the presence of host than one

stimuli received alone. It is also possible that horn fly

pheromones may influence attack of host (Milstrey, 1983).

Horn flies are herd or individual animal parasites,

therefore any effect which keeps them on the same host

regulates the host parasite interaction.

Understanding all of these factors influencing insect

infestation will require the development of new research

tools. Successful identification of the factors conferring

refractiveness by chemical augmentation, host selection or

semiochemical activation could help in the breeding of more

resistant animals as part of an integrated control regime.

This could reduce cost by reducing the use of chemicals. To

determine specifically how the attractancy or repellency of

the washes affects the behavior of the flies, further

bioassays are needed because experimental design used in

bioassay may restrict the nature of stimulus change and fly

response. Follow-up field studies are important in

confirming laboratory observations. Field and laboratory

techniques sometimes give different indications of activity

which can be misleading as, for example the range of

distance at which flies can detect the washes.

Insecticide impregnated baits made of blends of

acetone, l-octen-3-ol, 4-methylphenol and 3-n-propylphenol

are used in Tsetse flies (Glossina spp) control. This

control strategy was based on an understanding of the

responses of tsetse flies to their host, using research

tools that quantify single specific responses (Torr, 1990).

Humphreys and Turner (1973) reported that host size is a

factor in attracting biting midges Culicoides, but that host

color not significant in attracting two species of this


Knowledge of factors conferring repellency to hosts

could help to protect man and animals against vector-borne

diseases. Animals exhibit strategies to avoid, control or

eliminate parasites. These strategies are broadly divided

into 5 groups (Hart, 1990). One describes behaviors

allowing animals to avoid or minimize exposure to

arthropods. The second group control exposure time during

which they and their offspring are exposed to parasites.

They are thus exposed to small doses of parasites and this

helps in the development of immunity. The third group

depends on behavioral patterns of anorexia and depression.

These are observed when animals get sick with febrile

diseases. This potentiates the fever responses enabling the

animal to overcome infection. The fourth strategy relates

to the tendency of some animals to help group mates or kin

that are sick or injured. The fifth strategy is that in

which animals select mates on the basis of evidence of

resistance to parasite infection thus producing offspring

with the genetic basis for resisting parasites. Thus the

factors conferring refractiveness or resistance of animals

are varied, and meticulous studies are needed to identify

the most important and influential factors. Currently the

innate resistance of cattle to biting arthropods has been

masked for several decades by the use of insecticides which

have allowed genetically susceptible cows to compete

favorably within the herd (Steelman et al., 1993), making it

difficult for research into the most important resistance

factors. Knowledge has been gained of the behavioral

responses of the horn fly to different host odors with this

study. The results obtained provides a good starting point

for measuring host resistance and for testing chemicals and




The horn fly is one of the most important pest of

cattle in the United States (Marlatt, 1910; Dosey et al.,

1962; Bruce, 1964). Losses caused by the fly, the

increasing resistance of the fly to the action of

insecticides and the high cost of chemicals have stimulated

research into alternative control measures of which the

possible use of anti-arthropod vaccine is part. Vaccine

development will require accurate knowledge of naturally and

artificially induced immunity. It is envisaged that

antibodies produced in a host by natural or artificial

immunization could attach to a wide range of potential

targets within an insect, disrupting selected tissues and

physiological processes. Wikel (1983) and Willadsen (1980)

listed premature detachment, reduced engorgement size,

increased mortality, decreased fecundity and diminished

hatching in ticks as some of the consequences of host



Blood feeding ectoparasites present a limited range of

antigens to the host with introduction of their saliva

(Nelson et al., 1977). The host's immune system,

recognizing these antigens in the saliva, produces

antibodies to counteract them. When the antibodies are

taken in the blood meal by hematophagous insects, they bind

to their antigenic counterparts within the insect, and may

disrupt cellular function and increase mortality. Host

antibodies interfere with normal tick feeding by inhibiting

the activities of secreted salivary enzymes (Reich and

Zorzopulus, 1980). Anti-tick vaccine has been produced and

successfully used to protect cows. A variety of extracts

from tissues which normally have no contact with the host

induce immune protection to tick infestation (Galun, 1975b;

Ackerman et al., 1980; Mongi et al., 1986; Willadsen, 1987).

Identification and characterization of appropriate

antigen(s) and the mechanism of presenting these antigens to

the host in a manner that stimulates the proper long lasting

immune response is important in the development of a

vaccine. Identification of antigens in blood-feeding

insects requires knowledge of the functions of saliva, since

it is often introduced into host. There is little knowledge

of the function of saliva in horn flies and many other

important ectoparasites.

Various roles have been ascribed to the saliva of

hematophagous arthropods, though until recently no general


role for saliva in hematophagy was apparent (Ribeiro, 1987).

Some blood-feeding insects feed in the absence of saliva

(Lester and Lloyd, 1928; Hudson et al., 1960; Hudson, 1964;

Rossignol and Spielman, 1982). Saliva also functions in

ways unrelated to blood-feeding (Ribeiro, 1987). It is

involved in dissolution of solid sugar (Eliason, 1963) and

in lubrication of the feeding stylets in mosquitoes (Orr et

al., 1961). Gooding (1972) found anticoagulants and

hemagglutinins in the saliva or salivary gland homogenates

of some blood-feeding arthropods but could not ascribe a

role to them. Kerlin and Hughes (1992) reported that

salivary secretions from parasites probably were produced to

promote parasite feeding and survival on the host. Saliva

from Ixodes damini contain anti-haemostatic,

immunosuppressive, anti-inflammatory, and neutrophil-

inhibiting components (Ribeiro et al., 1985). Saliva of

some hard ticks contain prostaglandins that may help feeding

by increasing host skin circulation at the site of the bite

(Binnington and Kemp, 1980; Kemp et al., 1982). Their

salivary apparatus also contributes to ion and water

metabolism by excreting excess water acquired from the blood

meal (Binnington et al., 1980; Kemp et al., 1982).

Saliva from blood-sucking insects contains toxins or

allergens (Gothe et al., 1979; Wikel, 1982; Wikel and Allen,

1982). Host allergic or hypersensitive response to saliva

introduction functions to diminish arthropod feeding success


(Wikel and Allen, 1982). Repeated exposure to salivary

antigens may cause a host to produce antibodies that may

alter the feeding site of the arthropod (Wikel, 1982; Wikel

and Allen, 1982).

Any vaccine development must consider several

questions. These include the type of effector cell involved

in resistance, the protective antigens that are recognized,

and the form in which these exogenous proteins are


The purpose of this work is to identify and

characterize antigens of potential value. This study

attempts to identify antigens from the salivary gland and

salivary secretion of the horn fly using Sodium dodecyl

sulphate polyacrylamide gel electrophoresis (SDS-PAGE).

Western blots of SGE were reacted with rabbit anti-salivary

gland extract and mouse anti-salivary secretion serum to

identify proteins bands of immunological importance.

Materials and Methods

Horn Fly Rearing And Culture Procedures

Horn fly (Haematobia irritans (L.)) adults were

obtained from the laboratory colony at the University of

Florida (Greer, 1975; Bolton, 1980). Standard colony

rearing methods were modified from those of Greer (1975) as

follows: A larval medium was made from a mixture of 1370 g

of cattle manure (frozen at -20C for at least 48 hours to

kill all naturally occurring dung parasites), 546 g


pelletized peanut hulls (High Springs Milling Co Ltd., High

Springs, FL.) and 258 g dry mix (44% all-purpose flour, 33%

fish meal, 18% alfalfa meal, and 5% baking soda). Peanut

hulls were presoaked in 700 ml of water for 30 minutes, and

then crushed by hand. The dry mix was added to the peanut

hulls and mixed thoroughly before the manure was mixed in.

Horn fly eggs (0.5 ml) were added to about 1400 g of medium,

which was maintained for 7 days in an environmental chamber

(PercivalR Boone, Iowa) at 27 3C and 75 5% R.H., with

continuous light supplied by two fluorescent tubes. Pupae

were separated from the medium by water floatation and dried

on paper toweling. Dry pupae were held in a screen-covered

aluminum cage (51 X 26 X 27 cm) in an environmental chamber

until adults closed.

Adults were fed whole bovine blood obtained from

freshly slaughtered cows. One liter of blood was collected

in a 2.5 liter vessel containing 3.96 g of sodium citrate in

13.2 ml of water. KantrexR powder (Apothecon, Princeton,

NJ) (0.24 g) and 0.022 g of Mycostatin R (E. R. Squibb &

Sons, Inc. Princeton, NJ) powder were added to 1 liter of

citrated bovine blood to control microorganisms and prevent

spoilage. The blood was stored at about 4C until needed.

Beef juice, the fluid drawn off from stored cut or ground

meat was obtained locally and was added to the citrated

bovine blood at a ratio of 75 parts blood:l part beef juice.


Dissection Of Horn Fly Salivary Glands For Antigen

Salivary glands were surgically removed by the

following protocol. Groups of general adult flies (24-48

hours old) were immobilized by cooling for about 20 minutes

at 1C. The immobilized flies were glued dorsally with

Super GlueR onto wax strips demarcated 3 X 1 inch glass

slides. The flies were flooded with insect ringers solution

(Smith and Butler, 1991). Legs were first removed with a

fine tipped forceps, the ventral integument was then removed

anterior-posteriorly from the thorax to the abdomen. The

exposed salivary glands were teased out and removed with the

forceps (Fig. 3.1). The glands were washed and stored in

insect Ringers solution containing 10 mM of

Phosphonylmethyl-sulfonyl fluoridate (PMSF) kept on ice.

PMSF phosphorylates the active site of serine proteases and

inhibits their activity. The salivary glands were

homogenized in a Tenbroek tissue homogenizer, sonicated

(55,000 cycles/second) in 30 second bursts on ice and

centrifuged at 13,000 g for 20 minutes to remove insoluble

material. The supernatant was decanted and was referred to

as salivary gland extract (SGE). The SGE was divided and

stored at -20C until used. The concentration of protein in

the SGE sample was determined by the Bradford method

(Bradford, 1976) with bovine serum albumin as standard.


Polyclonal Antibody Production.

Two New Zealand female white rabbits (3-4 months old)

and two female ICR mice were used as models to generate

antibodies against horn fly salivary components. The mice

and rabbits were kept in individual cages fed a pelleted

diet and water ad libitum, and kept on a 12:12 light-dark

cycle. Care was taken to prevent ectoparasitic infection of

the animals.

Two different methods utilizing two antigen sources and

two different host models were used for antibody production.

These were: (1) Immunization of rabbits with SGE with

Freund's adjuvant and (2) Naturally infesting caged

restrained mice (Fig. 3.2) by exposing these mice to horn

fly bites. By this method salivary secretions associated

with horn fly blood feeding served as the immunogen.

(1) The immunization with the crude SGE was contracted

to Kel, Inc. Gainesville, Florida, who performed the

following immunization and bleeding protocols. Preimmune

blood samples from the unimmunized animals were taken prior

to immunizations. A volume (1ml) of SGE antigen (2 mg/ml),

was emulsified with an equal volume of Freund's complete

adjuvant and injected subcutaneosly into each of the

rabbits. This was followed by bi-weekly immunizations of

antigens in Freund's incomplete adjuvant. On day thirty,

sixty and ninety post immunization (PI) blood samples were

collected by cardiac puncture from the rabbits. The


collected blood was clotted overnight. Serum was separated

by centrifugation, decanted, pooled and stored in small

aliquots at -20C until needed.

(2) Preimmune serum was obtained from the mice before

infestation. Two mice were cage restrained (Fig. 3.2) and

sequentially exposed twice weekly to (24-48 hr old) general

horn flies. The immobilized mice were placed in a cage

containing about 100 horn flies of both sexes. On days 30,

60 and 90 antisera were obtained from the mice by cardiac

puncture. The blood was allowed to clot. The clotted blood

was then centrifuged at 20,000 g for 15 minutes and serum

obtained. The serum from each experimental group was

pooled. The pooled antisera was stored in aliquots at -20C

until used. The sera from the two model animals were tested

for antibody production with ELISA. When a high antibody

titer had been obtained animals were sacrificed and antisera

was obtained for tests.

DOT-Enzyme-Linked Immunosorbent Assay (DOT-ELISA)
For Determining Optimum Antigen Titer.

Assay was performed using serial dilutions of the SGE

stock solution in PBS containing from 1 ug/ml to 20 ug/ml.

The serially diluted (2 ul) samples were pipetted

individually onto spots on two (20 X 10 cm) PVDF

(Polyvinylidene difluoride) protein sequencing membranes.

The PVDF membrane had previously been presoaked with

absolute methanol. Two dots of PBS/T were used as a


negative control and another two dots which had antigen but

treated with normal mouse serum (NMS) and normal rabbit

serum (NRS) were used as positive control. The PVDF

membranes were handled with gloved hands and forceps to

prevent contamination. The dotted PVDF membranes were

immersed in 3% BSA-PBS-T (3% bovine serum albumin-0.01 M

phosphate buffered saline containing Tween 20), washed in

PBS-T (phosphate buffered saline containing Tween 20) drying

agents and air dried. Two microliters of serially diluted

rabbit anti-SGE serum and mouse anti-salivary secretion

serum (1:50 to 1:3200) were applied on top of the dots on

the two membranes, respectively. After 30 min incubation,

the PVDF membrane were washed 3-5 times in PBS-T, soaked in

alkaline phosphatase-conjugated goat anti-rabbit IgG or goat

anti-mouse IgG at 1:1000 dilution, respectively, and left to

react at room temperature for 30 minutes. The PVDF

membranes were washed three times in PBS-T, once in PBS, and

immersed in substrate solution, freshly prepared by mixing

equal volumes of BCIP and NBT. After color development the

PVDF membrane was then washed with water and air dried. The

intensity of the yellow color was visually compared in

reference to the negative and positive controls.

Antibody Titer Determination Using Enzyme-Linked
Immunosorbent Assay (ELISA)

An ELISA was used to determine the titer of model

animals antibody response to immunization with SGE and

natural horn fly infestation. To determine optimum antibody

titers, pooled serum from the rabbits immunized with SGE and

naturally exposed mice were assayed on days 30, 60 and 90.

One hundred uls of a working dilution of 2 ug/ml per

well (determined as optimum by DOT blot ELISA assay) of SGE

diluted in PBS (pH 9.5) was placed in a 96 well flat-

bottomed Nunc Maxi-sorp, immunoassay plates (Dynatech,

Chantilly, VA) to coat the plates. Positive controls

containing antigen but reacted with NMS and NRS and negative

controls containing (PBS) were run simultaneously. The

negative control was used to give background optical

density. The plate was incubated overnight at 4C.

Following binding of the antigen to the plates, the plates

were washed 4X with wash buffer (PBS with 0.02% sodium

azide, 0.05% Tween 20). The non-specific binding sites were

blocked by adding 300 ul/well of blocking buffer (1% BSA in

PBS with 0.02% Sodium azide) to the microtiter plate and

incubating the plate for 60 minutes at room temperature.

These were washed 4X with wash buffer as described above.

Wells were then incubated with 100 ul/well of serially

diluted (1:50 to 1:3200) rabbit anti-SGE, and mouse anti-

salivary secretion serum. The wells were incubated for 60

minutes at room temperature and were then washed with wash

buffer as described previously. One hundred ul/well of

rabbit anti-mouse IgG whole molecule alkaline phosphatase

conjugated (Sigma) diluted (1:1000) in PBS with 0.02% azide,


and goat anti-rabbit IgG whole molecule alkaline phosphatase

conjugated (1:1000) were added to the wells containing the

respective primary antibodies as secondary antibody and

incubated at room temperature for 60 minutes. The wells

were then washed as described above. One hundred ul/well of

1 mg/ml freshly made substrate, Para-nitrophenyl phosphate

(Sigma) in diethanolamine substrate buffer (Sigma Chemical

Co., St. Louis. MO) with 1 mM MgCl2. pH 9.8) was added to

the plates and incubated for 60 minutes at room temperature

in the dark. Color development was stopped by the addition

of 50 ul/well of 3M NaOH. The absorbance was read at 405 nM

at 30 and 60 minute time intervals on a Dynatech Micro ELISA

reader (Dynatech Laboratories).

Protein Separation With Polyacrylamide Gel Electrophoresis

SGE antigens and known protein standards were

electrophoresed on sodium dodecyl sulphate-polyacrylamide

gel electrophoresis (SDS-PAGE) using 12% separating gel

preparation with 4% stacking gels (Bio-Rad Laboratories)

(Laemmli, 1970) Protein bands were visualized with

coommassie brilliant blue stain. The molecular weight of

separated proteins was estimated by comparison with

electrophretically separated low molecular weight and high

molecular standards (Bio-Rad laboratories, 32nd and Griffin

Ave., Richmond, Calif.).

Figure 3.1. Dissection of the salivary glands of the horn

Figure 3.2. Restrained mouse ready to be offered to horn
flies for feeding


Twenty (20) ul of SGE sample containing 5-20 ug/ml of

soluble protein were diluted in equal volume of 2X sample

buffer (0.08 M Tris-HCL, pH 6.8, 0.1 M dithiothritol, 2%

SDS, 10% glycerol, 10% 2-mercaptoethanol, 0.2% bromophenol

blue) and heated in a water bath at 100C for 4 min. These

prepared samples were loaded into the sample wells (10-20

ul). Molecular weight of electrophoretically separated

protein bands were estimated by measuring the migration

distance of protein bands against prestained low molecular

weight standards (Phosphorylase b, Mr=130KDa; Bovine serum

albumin (BSA), Mr=75KDa; Ovalbumin, Mr=50KDa; Carbonic

anhydrase, Mr=39KDa; Soyabean trypsin inhibitor, Mr=27KDa

and Lysozyme, Mr=17KDa) included with each electrophoresis

run. Bromophenol blue incorporated in the buffer allowed

monitoring of protein migration during electrophoresis. Two

slab gels were run concurrently. For optimum protein

separation with 2 gels, voltage was held constant at 200 v

until the sample migrated into the separating gel. When the

bromophenol blue dye front was about 2 mm from bottom of the

gel the voltage was shut off. Gels were fixed and protein

bands were visualized by staining with 0.1% coomassie

brilliant blue R-250/40% methanol/acetic acid for 2 hours

and destined with 40% methanol/10% acetic acid.

Western Blot

Following SDS-PAGE of the SGE, the unfixed resolved

proteins were transferred electrophoretically onto PVDF

Immobilon transfer membrane (Millipore Corp. Bedford, Mass)

employing the method of Towbin et al. (1979). Proteins were

transferred from the gel to the PVDF protein sequencing

membrane at constant voltage (25 V) overnight at room

temperature. The gel and PVDF blot were removed from the

cassette once the transfer was complete. The gel was

stained with 1% coomassie blue to ascertain if the proteins

were adequately transferred. The PVDF protein sequencing

membrane were equilibrated in Tris buffered saline (TBS) for

30 minutes on a rocker and, then, blocked in a solution

containing bovine serum albumin and fetal calf serum (3%

BSA, 5% fetal calf serum, 0.1% Tris buffered saline with

Tween 20) for one hour. The PVDF membrane were transferred

to TTBS for 2 successive washes (5 mins each) and air dried.

It was then put in a zip lock bag and frozen at -20oC until

used. The PVDF blot was cut into 1 cm vertical strips

corresponding to the individual lanes of migrating protein.

Immunodetection Of Salivary Gland Antigens

The affinity of the pooled antisera from the two model

animals for the electrophoretically separated and

immobilized SGE antigens was tested. The methodology used

was a modification of that described by Towbin et al.


(1979). Two strips of the blotted PVDF membrane were socked

in bovine serum albumin and fetal calf serum block for one

hour. This was then washed 3 times with TTBS for about 5

minutes each. The two strips were then incubated either

with the polyclonal primary antibodies from the rabbit or

the mouse for 3 hours at 1:200 dilution in TBS (found to be

optimum with ELISA). The strips were then washed 3 times

for 5 minutes each in Tris buffered saline and 0.5% tween 20

(TTBS) to remove the unbound antiserum. The attached mouse

antibody was labelled by alkaline phospatase attached to

goat anti-mouse secondary antibody (1:1000 in block

dilution) and the rabbit antibody was labelled by alkaline

phosphatase attached to goat anti-rabbit secondary antibody

for 1-2 hours. These were used to visualize the SDS-PAGE

separated SGE proteins. The blots were removed and washed

3X for 5 minutes each with TTBS. Color development was

achieved by placing the blots in 0.1 M Tris-HCl buffer, 10

mM MgCl2 pH 8.8 for 15 minutes. This was decanted and the

substrate BCIP/NBT (5-bromo-4-chloro-3-indoxyl phosphate and

nitrozolium blue (sigma) (50 ml of 0.1 Tis/MgCl2 solution pH

8.8, 50 ul of BCIP and 100 ul NBT) was added. Color

development was stopped by immersion in distilled water for

two 10 minutes washes. The strips were dried between filter

paper and photographed.


Evaluation Of SGE For Carbohydrate Epitopes

Periodate oxidation reaction was used with ELISA

technique to evaluate the reaction of the mouse and rabbit

polyclonal antibodies to SGE as to reaction to carbohydrate

or non-carbohydrate determinants (Woodward et al., 1985).

To identify the type of epitope found in SGE, ELISA titers

of periodate treated and non-treated SGE were compared. Two

microtiter plates were used, one for the periodate assay and

the other used as control. Fifty microliters of a working

dilution of 2 ul/ml of SGE diluted in PBS (pH 9.5) was added

to the microtiter plates. The plates were incubated

overnight at 4C. The plates were then washed 4X with wash

buffer (IX pbs with 0.02% sodium azide, 0.05% Tween 20).

Non-specific binding sites were blocked by adding a blocking

buffer 1% BSA (Bovine Serum Albumin) in PBS with 0.02%

sodium azide and incubating in block for 60 minutes at room

temperature. The wells were then washed with the buffer

described above. Each microtiter plates was then washed

with 50 mM sodium acetate buffer (pH 4.5). One microtiter

plate was incubated with 100 ul/well of 10 mM sodium meta-

periodate (Sigma Co. St. Louis, Mo) in 50 mM sodium acetate

buffer, pH 4.5, for an hour at 25C in the dark. Following

a brief rinse with 50 mM sodium acetate both plates were

incubated with 100 ul/well 5 mM sodium borohydride (Fisher

Scientific, Fairlawn, NJ) in PBS for 30 minutes at 25C.

The reagents used were freshly prepared just before use.


After 4 washes with wash buffer the ELISA procedures were

performed as described previously.


Optimum Antigen And Antibody Titers

An optimum concentration of 2 ug/ml of SGE per dot

allowed detection by both polyclonal antibodies at 1:200

serum dilution. This concentration of antigen was therefore

used to coat ELISA plates. ELISA reading of antisera from

immunized rabbits and ICR mice showed that high antibody

production was stimulated about 60 days after immunization.

The results of the ELISA experiment with pooled mouse and

rabbit antisera at day 60 post immunization and baseline

serum are shown in Fig. 3.3. With positive controls

(antigen reacted with Normal mouse serum (NMS) and normal

rabbit serum (NRS)) as base line, the titer of the anti-SGE

response and the mouse anti-salivary secretion response are

indicated as the last dilution of sera with an optical

density of 0.05 higher than the base line. Mean OD for

rabbits ranged from 0.5 to 2.25. That for mice ranged from

0.25 to 0.5. The optical density decreased with increasing

dilution of the sera samples. Low absorbance values (less

than 0.05) were obtained with the negative controls. The

titer of both the mouse and rabbit antisera was determined

to be 1/200 dilution. At high dilutions, antibody was still

detected by the ELISA procedures (1/6400 by the rabbit

antisera and 1/3200 for mouse antisera). There was a higher

antisera titer in the rabbit compared to the mice. From

these results it can be seen that all infected animals

showed antibody titer significantly (p=0.05) above that

found in uninfected animals and the antibody generation

between the two procedures was also significant.

Polyacrylamide Gel Electrophoresis. Western Blot And
Characterization Of Salivary Gland Aantigens

Denaturing, discontinuous SDS-PAGE analysis as

described by Laemli (1970) was used to analyze the proteins

in the SGE from the horn fly. The SGE separated into about

12 bands (Fig. 3.4). Readily detected proteins range in

molecular weight from about 75 KDa to less than 14 KDa.

Three prominent or major bands were observed consistently at

about 27, 50 and 75 KDa. Western blot analysis of the SGE

using the pooled diluted (1/200) sera from the rabbits and

the mouse show that the mouse antisera recognized 2 protein

bands, one major one and a minor one at about 27 KDa (Fig.

3.5). Seven reactive bands were precipitated by the rabbit

anti-SGE. A major band was recognized just below 30 KDa.

Two bands were just below 21.5 KDa, three bands between 46

and 69 KDa and a band at about 200 KDa. Normal rabbit and

mouse serum had no reactivity to salivary gland antigen

(Fig. 3.6 and 3.8 respectively). The proteins were

characterized according to molecular weight by comparing

their relative mobilities on SDS-PAGE with those of known


molecular weight standards. Both sets of sera consistently

reacted with the major protein band at about 27 KDa. This

was therefore identified as a secretary protein.

Periodate Oxidation

Antibody binding of the SGE antigen treated with meta-

periodate was reduced by sera from both host models. The

reduced antigen-antibody reactivity resulted from the

oxidation of the carbohydrate epitopes by the sodium meta-

periodate. The OD level of the rabbit antibody reaction was

reduced by a greater percentage by the periodate reaction

compared to the reduction of the mouse antibody reaction

(Table 3.1). This results shows that the polyclonal

antibodies recognized carbohydrate-conjugated epitopes

associated with glycoproteins making up the SGE antigen.


The need for adequate quantities of high quality

antiserum in immunological investigations cannot be

underestimated. The two procedures (immunization and

natural infestation) used for the antibody generation in the

model animals enabled a response to both the potential and

"true" salivary antigens to be compared (Cross et al.,

1993). Using the procedures also overcome the difficulty in

obtaining pure and adequate horn fly salivary antigen for



-4- -
- --





50 100 200 400 800 1600 3200 6400

Figure 3.3. Elisa showing binding efficiency of rabbit
anti-SGE and mouse anti-salivary secretion sera. Secondary
antibodies were coupled to alkaline phospatase and presence
of bound antibody was detected colormetrically using p-
nitrophenylphosphate as a substrate.
(representative data from 3 replicate experiments are


2 3

130 KDa
75 KDa
50 KDa
39 KDa
27 KDa


Figure 3.4. Photomicrograph of 12% SDS gel, depicting the
protein profiles of salivary gland extract from the horn fly
stained with coomassie blue. Lanes 1 and 2 show the
proteins in the SGE. Lane 3 shows the molecular weight

\ .. i. .


130 KDa-

27 KDa


Figure 3.5. Western blot of SGE exposed to mouse anti-
salivary secretion antibody and visualized with alkaline
phospatase labelled secondary antibody. The mouse serum
precipitates a broad band at approximately 27 KDa. Lane 1
and 2 the immunoprecipitated bands and lane 3 shows the
molecular weight markers.

200 KDa

4 97.4 KDa

j 69 KDa

46 KDa

1 30 KDa

21.5 KDa
14.3 KDa

Figure 3.6. Immunoblot of SGE exposed to NMS (Normal mouse
serum). Note the lack of immunoprecipitation.

1 2


97.4 KDa

69 KDa


30 KDa




Figure 3.7: Immunoblot of SGE exposed to rabbit anti-SGE
serum. Lane 1 shows the molecular weight markers. Lane 2
shows the precipitated bands. Note the broad band just
below 30 KDa.

200 KDa
97.4 KDa

e KDs

46 KDa

30 KDa

21.5 KDa
14.3 KDa

Figure 3.8. Immunoblot of SGE exposed to NRS (Normal rabbit
serum). Note the absence of any precipitation.

Table 3.1.

Optical Densities Of Periodate Treated SGE And
Binding Of Mouse And Rabbit Antibody.

Treatment Antibody OD (405)a

Controlb Rabbit anti-SGE 2.06

SGE + periodate Rabbit anti-SGE 0.13

Control Mouse anti-SSc 0.61

SGE + periodate Mouse anti-SS 0.03

aResults were averages from 3 readings.
bThe controls were treated with buffer (acetic acid-sodium
acetate at pH 4.50 followed by sodium borohydride.
c Antibody from mice against the salivary secretion
associated with horn fly feeding on mice.


immunological studies, it also provided a basis for

comparing the effect of the two procedures. The results of

the ELISA (Fig. 3.3) showed that the model hosts developed

an immune response to antigens from the salivary glands of

horn flies. This is shown by the development of high

antibody levels in the animals. Consistent increases in

concentration of immunoglobulin was observed for both model

animals but different antibody levels were obtained for

both. The high antibody titer observed with the ELISA with

the two model animals showed that the two laboratory animals

could be used for the study of host immune response to horn

fly salivary antigens. The lower antibody levels developed

in the mice exposed to horn fly feeding could be attributed

to the horn flies not producing enough critical mass antigen

to elicit adequate antibodies. The differences in the

antibody levels generated in the model hosts may reflect

variations in the response of the two different hosts to the

different antigens and also the different modes of

immunizations employed. Ribeiro (1989) reported differences

in relative amounts of anaphylactic response mediators among

vertebrate hosts. Despite the low antibody titer of the

mouse, it was able to immunoprecipitate protein bands later

on. One advantage of a high antibody titer in immunological

studies is that it allows for a high antisera dilution which

dilutes out population of unwanted or nonspecific antibodies

that might interfere in immunological work. This allows for


efficient use of antisera and allow for the fullest use of

the available quantity of high antibody, which often is

expensive and difficult to produce (Polak and Van Noorden,

1984). One of the most important attributes of good quality

antibody is its high affinity for its antigen, its binding

sites should fit well with the antigenic sites on its

specific antigen and not attach to other antigens (Polak and

Van Noorden, 1984). The affinity of the SGE and the

antisera from the two model animals seems good but the

possibility exists to obtain better affinity with pure horn

fly salivary antigen. This cannot be presently compared

because of the failure to obtain adequate quantities of horn

fly salivary secretion. To enhance antibody production

intra splenic immunization or in vitro methods could be

used. Different methods have been employed to obtain saliva

from arthropods. Oral secretions of fleas has been

collected directly into containers (Michaeli et al., 1966),

those of mosquitoes into distilled water (Allen and West,

1966), or those of ticks into capillary tubes (Clarke and

Hewetson, 1971). The buffalo fly, Haematobia irritants

exigua was induced to salivate with 80 mM serotonin (5-

hydroxytryptamine) in saline but very minute quantities were

obtained. Even with pure antigens, the antiserum produced

cannot be guaranteed to be directed specifically and solely

to the injected antigen (Polak and Van Noorden, 1984). The

antibodies produced could be directed to various parts of


the antigen molecule and to the carrier protein, or part of

it. Elisa provides a good screening test for antibodies and

can be used to check for cross reactivity or contaminating

antibodies (Polak and Van Nooren, 1984).

In the development of an anti-arthropod vaccine, an

understanding of the structure and function of the antigen

and the mechanism of the host response are essential. The

use of vaccines made up of antigens from the saliva and

other tissues has been shown to be effective in controlling

ticks (Wikel et al., 1992; Agbede and Kemp, 1986). The

possibility of exploiting the host immune response to

control flies was investigated by Schlein and Lewis (1976).

They demonstrated that hematophagous flies exhibit increased

mortality and cuticular abnormalities when allowed to feed

on rabbits immunized with fly extracts or fly integuments or

other tissues. Detailed studies of the horn fly salivary

antigen and proteins that can induce immunity in host are

essential processes in the development of an anti-arthropod

vaccine. The results of the SDS-PAGE demonstrated that SGE

is made up of multiple proteins (Fig. 3.4). Probing of the

western blot of the SGE with the mouse antisera precipitated

two bands (Fig. 3.5), and the rabbit antiserum precipitated

7 bands (Fig. 3.7). The consistent identification of a

common band at about 27 KDa by both antisera leads to the

conclusion that the protein is secretary and warrants

detailed studies for its suitability as a potential antigen


for vaccine production. The immunoprecipitation of fewer

protein bands by antisera from the naturally exposed mouse

as opposed to the immunized rabbit is not unusual. Similar

results were observed in work with tick salivary antigen and

black fly salivary antigens (Cross et al., 1993; Nyindo et

al., 1989). The results probably reflect variations in the

antigen used and the mode of immunization. Variation in the

results is important because it may be due to low level of

expression of the salivary secretion when introduced by the

horn fly as opposed to the high level of expression when

introduced through immunization. The identification of one

protein band as secretary antigen may also indicate that

only few antigens are present in the salivary secretion of

the horn fly or that few antigens in the saliva generates an

immune response. It has been observed in other studies that

antibodies were not detected against molecules corresponding

to the molecular weight of substances known to be

pharmacologically active in black fly saliva (Jacobs et al.,

1990; Wirtz, 1990). This indicates that the host does not

mount a response which would potentially interfere with the

functions of these components (Cross et al., 1993). The

periodate study indicates that the SGE possesses

carbohydrate antigenic determinants which may be involved in

antibody binding.

Understanding of the immune reaction of immunized hosts

and naturally exposed host could help in the identification


of antigens to be used in a vaccine and the form in which it

should be presented to host to obtain resistance. The

likelihood that the horn fly salivary antigen changes during

feeding makes the study of salivary heterogeneity important.

Heterogeneity of salivary gland secretion during feeding has

been observed in ticks. The size, mass and protein content

of the salivary gland increased about 25 fold during feeding

(McSwain et al., 1982). Kerlin and Allingham (1992)

demonstrated that cows do not exhibit immunity to buffalo

fly attack even when high levels of antibodies to buffalo

fly salivary antigens have been recorded. They concluded

that buffalo fly salivary antigens could not be good

candidates for vaccine production. On the contrary it has

been demonstrated that tick anti-saliva antigen imparted

immunity onto the host (Brown and Askenase, 1986; Shapiro et

al., 1989; Jaworski et al., 1990). Nelson et al. (1977)

reported that the active fraction in an extract injected may

be infinitesimal, or may require a complementary substance

to induce sensitization. Thus the host could not possibly

be able to detect all the antigens in the saliva, since it

is possible that only a small fraction of the salivary

components are active. The preparation of the SGE antigen

may also affect results obtained. This is because the

temperature requirement for active secretion and action of

the saliva may be incompatible with processing and

preservation procedures used in this study. The insect may


also secrete a substance that cannot be found in the

extract. The seven bands identified by the rabbit sera

could be potential antigens and further study is required to

elucidate this. The presence of the homologous 27 KDa band

shows that at least some of the active ingredients in the

saliva had been captured by the SGE preparation.

The identification of antigens from the saliva of the

horn fly using rabbits and mice as models was based on the

concept that immunizing hosts with vaccines composed of

those antigens would elicit antibodies, which when ingested

by the fly would bind to target tissues, producing

deleterious effects. Tick salivary gland immunogens are

responsible for stimulating and interacting with the host

immune response. The lack of any demonstrated protective

function for host immunoglobulin against many hematophagous

insects indicates the need for further research to produce

anti-arthropod vaccine.

Salivary antigens may be poor antigens. The horn fly

also may have the ability to evade or depress the host's

immune system to ensure feeding success. Wikel and Whelen

(1976) and Barriga et al. (1991) demonstrated that tick

infestation depresses the host's immune response. This is

done in part by inhibiting the production of effective T

cell or specific antibodies. Horn flies produce little

quantities of saliva as compared to stable flies or ticks

(Butler et al., 1977). They also reported that as a result