Louisiana State University LSU Digital Commons
LSU Historical Dissertations and Theses Graduate School
1988 Antigenic Analysis of Equine Infectious Anemia Virus Using Monoclonal Antibodies. Khalid Abdullah Hussain Louisiana State University and Agricultural & Mechanical College
Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses
Recommended Citation Hussain, Khalid Abdullah, "Antigenic Analysis of Equine Infectious Anemia Virus Using Monoclonal Antibodies." (1988). LSU Historical Dissertations and Theses. 4508. https://digitalcommons.lsu.edu/gradschool_disstheses/4508
This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS
The most advanced technology has been used to photo graph and reproduce this manuscript from the microfilm master. UMI films the original text directly from the copy submitted. Thus, some dissertation copies are in typewriter face, while others may be from a computer printer.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyrighted material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are re produced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each oversize page is available as one exposure on a standard 35 mm slide or as a 17" x 23" black and white photographic print for an additional charge.
Photographs included in the original manuscript have been reproduced xerographically in this copy. 35 mm slides or 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
Accessing theUMI World’s Information since 1938 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA
Order Number 8819948
Antigenic analysis of equine infectious anemia virus using monoclonal antibodies
Hussain, Khalid Abdullah, Ph.D.
The Louisiana State University and Agricultural and Mechanical Col., 1988
UMI 300 N. ZeebRd. Ann Arbor, MI 48106
PLEASE NOTE:
In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark -/ .
1. Glossy photographs or pages \S
2. Colored illustrations, paper or print ______
3. Photographs with dark background ____
4. Illustrations are poor copy ______
5. Pages with black marks, not original copy ______
6. Print shows through as there is text on both sides of p ag e ______
7. Indistinct, broken or small print on several pages ___
8. Print exceeds margin requirements _____
9. Tightly bound copy with print lost in spine ______
10. Computer printout pages with indistinct print ______
11. Page(s)______lacking when material received, and not available from school or author.
12. Page(s)______seem to be missing in numbering only as text follows.
13. Two pages num bered . Text follows.
14. Curling and wrinkled pages _
15. Dissertation contains pages with print at a slant, filmed as received ______
16. Other ______
UMI
ANTIGENIC ANALYSIS OF EQUINE INFECTIOUS ANEMIA VIRUS USING MONOCLONAL ANTIBODIES
A Dissertation
Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy
in
The Interdepartmental Program
in
Veterinary Medical Sciences with option in Veterinary Microbiology and Parasitology
by
Khalid Abdullah Hussain B.V.M. & S., University of Baghdad, 1977 M.S., Washington State University, 1983 May 1988 ACKNOWLEDGMENTS
I wish to express my sincere gratitude and appreciation to my major
professor, Dr. Charles Issel for his suggestions, constant support and
guidance with my research efforts and preparation of this dissertation.
Working with him has been rewarding and afforded me great opportunities
to interact and learn from other research groups around the country.
Thanks is also due to the members of my graduate committee, Drs. Kenneth
Schnorr, Ronald Montelaro, Thomas Klei, and Frederick Enright for their constructive advice and help. A special thanks is expressed to Drs.
Kenneth Schnorr and Ronald Montelaro for their assistance and expert support with several aspects of this research.
I am also thankful to W.V. Adams, J r ., Sue Hagius, Melanie West,
Linda Shaffer, Nancy Satterlee and Dr. Grace Amborski for their occasional assistance and support during this investigation. The assistance given to me by Jackie McManus and Tammy Allgood is greatly appreciated. I am grateful to the Ministry of Higher Education and
Scientific Research, IRAQ for it support through an academic scholarship and for the privilege to undertake advanced studies.
I am indebted to the Department of Veterinary Microbiology and
Parasitology of the School of Veterinary Medicine and the Department of
Veterinary Science, Louisiana State University, where this research was conducted. I wish to thank Drs. Johannes Storz and Kirklyn Kerr for their encouragement and help. Funds for this investigation were provided by the National
Institute of Health, the Louisiana State University School of
Veterinary Medicine, and the Louisiana Agricultural Experiment
Station, Louisiana State University Agricultural Center. To my family for your support, encouragement and patience.
To my wife, Souad. Without your constant love, endurance, and
understanding, my dream would never have come true.
To my son, Saad. Thank you for bringing more joy and love into my world than I could ever have imagined.
iv TABLE OF CONTENTS
Chapter Page
TITLE...... i
ACKNOWLEDGMENTS...... ii
DEDICATION...... iv
TABLE OF CONTENTS...... v
LIST OF TABLES...... ix
LIST OF FIGURES...... x
ABSTRACT ...... xii
I. INTRODUCTION ...... 1
A. LITERATURE REVIEW...... 1
1. Equine infectious anemia virus: etiology and clinical disease ...... 1
2. Viral characteristics ...... 3
a. Physicochemical properties ...... 3
b. Reverse transcriptase ...... 3
c. Morphology ...... 4
d. Genomic and antigenic classification. . . 4
e. Structural proteins and glycoproteins. . . 6
3. Virion surface glycoproteins ...... 8
a. Neutralization ...... 8
b. Use of monoclonal antibodies ...... 11
c. Antigenic variation ...... 17
4. Viral infected cell membraneantigens ...... 20
B. RESEARCH OBJECTIVES...... 21
II. GENERATION OF MONOCLONAL ANTIBODIES: PRODUCTION AND CHARACTERIZATION ...... 23
A. INTRODUCTION...... 23
v Chapter Page
B. MATERIALS AND METHODS...... 24
1. Cell culture...... 24
2. Virus propagation and purification ...... 24
3. Glycoprotein purification ...... 25
4. Immunization of mice ...... 26
5. Preparation of cells ...... 27
a. Myeloma cells SP2/0-Agl4 ...... 27
b. Feeder layer ...... 28
c. Splenic lymphocytes ...... 28
6. Hybridization ...... 29
7. Maintenance and specificity testing of hybridomas ...... 30
a. Care and selection of hybridomas ...... 30
b. Cloning by limiting dilution...... 31
c. Cryopreservation and thawing of clones. . . 31
8. Preparation of Ascitic fluid ...... 32
9. Determination of immunoglobulin class ...... 33
10. Screening of hybridomas ...... 33
11. Identification of viral proteins by the protein blot immunoassay "Western Blot" ...... 35
12. Preparation of reagents ...... 35
a. Growth media ...... 35
b. Polythylene glycol (PEG)...... 36
c. Hypoxanthine/Thymidine (HT) ...... 36
d. Ami no p te rin ...... 36
e. 8-Azaguanine ...... 36
C. RESULTS...... 37
v i Chapter Page
1. Enzyme-linked irranunosorbent assay ...... 37
2. Isolation of hybrid cells producting anti-EIAV antibodies ...... 37
3. Proteins recognized by sera from immunized m ic e ...... 40
4. Identification of viral proteins detected by monoclonal antibodies ...... 40
5. Characterization of monoclonal antibodies. . . . 42
6. Protein A binding assay ...... 42
D. DISCUSSION...... 48
III. ANTIGENIC ANALYSIS OF EQUINE INFECTIOUS ANEMIA VIRUS VARIANTS BY USING MONOCLONAL ANTIBODIES: EPITOPES OF GLYCOPROTEIN 90 (GP90) OF EIAV STIMULATE NEUTRALIZING ANTIBODIES...... 55
A. INTRODUCTION...... 55
B. MATERIALS AND METHODS...... 57
1. Virus production, propagation, and purification ...... 57
2. Enzyme linked immunosorbent assay (ELISA). . . . 58
3. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) ...... 60
4. Western blot immunoassay (immunoblotting). . . . 61
5. Neutralization of viral infectivity with monoclonal antibodies (Beta Procedure) ...... 62
6. Neutralization activity (Alpha Procedure). . . . 63
C. RESULTS...... 64
1. Western blot analysis ...... 64
2. Reactivity of monoclonal antibodies with the homologous virus stra in ...... 68
3. Reactions of monoclonal antibodies with a panel of EIAV v a r ia n ts ...... 68
v ii Chapter Page
4. Neutralization of EIAV infectivity by monoclonal antibodies ...... 76
5. Comparison of 16 EIAV i s o l a t e s ...... 82
D. DISCUSSION...... 89
IV. ANTIGENIC MAPPING OF THE ENVELOPEPROTEINS OF EQUINE INFECTIOUS ANEMIA VIRUS:IDENTIFICATION OF A NEUTRALIZATION DOMAIN AND A CONSERVED REGION ON GLYCOPROTEIN 90 ...... 104
A. INTRODUCTION...... 104
B. MATERIALS AND METHODS...... 105
1. Viruses...... 105
2. Glycoprotein purification...... 105
3. Production of monoclonal antibodies ...... 105
4. SDS-PAGE and Western blot assay ...... 106
5. Generation and detection of gp90 fragments . . . 106
6. Enzyme-linked immunosorbent assay -additivity t e s t ...... 106
7. Competition binding studies ...... 109
C. RESULTS...... 110
1. Epitope mapping ...... 110
2. Competitive ELISA ...... 116
3. Analysis of gp90 fragments ...... 116
D. DISCUSSION...... 120
V. SUMMARY...... 128
VI. BIBLIOGRAPHY...... 133
VII. APPENDIX...... 146
VIII. VITAE...... 148
v iii LIST OF TABLES
Table Page
11.1 Monoclonal antibodies to EIAV polypeptides ...... 43
111.1 Serum neutralization indices of three sera collected from pony #127 ...... 65
111.2 Characterization of anti-EIAV monoclonal antibodies ...... 69
111.3 Neutralization of EIAV by immune mouse serum and monoclonal antibodies ...... 80
111.4 Neutralization indices of positive horse serum and monoclonal a n tib o d ie s ...... 81
111.5 Classification of EIAV isolates according to reactivity pattern of gp90 or gp45 ...... 86
111.6 Serotyping of 16 EIAV isolates on the basis of reaction with MCAbs to gp90 and gp45 ...... 88
IV. 1 Additive enzyme-linked immunosorbent assay ...... 113
IV.2 Additivity index of gp90 specific monoclonal antibodies...... 114
IV.3 Additivity index of gp45 specific monoclonal antibodies ...... 115
IV.4 Summary of competitive ELISA ...... 118
ix LIST OF FIGURES
Page
Titrations in ELISA of immune mouse serum against EIAV antigen ...... 38
Hybridoma colony ...... 39
Western blot analysis of EIAV polypeptides. . . . 41
Western blot analysis of 10 anti-gp90 monoclonal antibodies ...... 44
Western blot analysis of 8 anti-gp45 monoclonal antibodies ...... 45
Western blot analysis of 5 anti-p26 monoclonal antibodies ...... 46
Immunodiffusion analysis ...... 47
Inoculation scheme followed for the production of EIAV isolates in experimental ponies ...... 59
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of seven EIAV isolates . 66
Western blot analysis of reference horse serum with seven EIAV iso lates ...... 67
Western blot analysis of gp90 specific monoclonal antibodies (86-1E3) with 16 EIAV isolates . . . . 71
Western blot analysis of gp90 specific monoclonal antibodies (87-IE7) with 16 EIAV isolates. . . 72
Western blot analysis of gp90 specific monoclonal antibodies (82-1C2) with 16 EIAV isolates . . . . 73
Western blot analysis of gp90 specific monoclonal antibodies (115-3D7) with 16 EIAV isolates. . . . 74
Western blot analysis of gp90 specific monoclonal antibodies (128-2B9) with 16 EIAV isolates. . . . 75
Western blot analysis of gp90 specfic monoclonal antibodies (85-1E11) with 16 EIAV isolates. . . . 77
Western blot analysis of gp45 specific monoclonal antibodies (105-3C8) with 16 EIAV isolates. . . . 78
Western blot analysis of gp45 specific monoclonal antibodies (90-1C1) with 16 EIAV isolates . . . . 79
x Figure Page
111.12 Western blot analysis of immune mouse serum and monocolonal antibodies with prototype EIAV and isolate P2.1-1 ...... 83
111.13 Immunoreactivity pattern of gp90- and gp45-specific monoclonal antibodies with polypeptides of 16 EIAV i s o l a t e s ...... 84
IV.1 Operational antigenic map of 16 EIAV isolates. . . . 107
IV.2 Saturation curves of EIAV with neutralizing monoclonal antibodies ...... 112
IV.3 Competitive inhibition of neutralizing monoclonal antibodies to EIAV glycoprotein in ELISA ...... 117
IV.4. Western blot analysis of natural and experimental EIA-positive horse sera ...... 119
IV.5 Western blot analysis of gp90 fragments of several serologically distinct EIAV isolates ...... 121
IV.6 Schematic model of the proposed topological arrangement of the epitopes on gp90 of EIAV ...... 123
xi ABSTRACT
ANTIGENIC ANALYSIS OF EQUINE INFECTIOUS ANEMIA
VIRUS USING MONOCLONAL ANTIBODIES
By
Khalid Abdullah Hussain
Monoclonal antibodies (MCAbs) produced against the prototype cell-adapted Wyoming strain of equine infectious anemia virus (EIAV), a lentivirus, were studied for reactivity with the homologous prototype and 16 heterologous isolates. Eighteen hybridomas producing MCAbs were isolated. Western blot (immunoblot) analyses indicated that 10 were specific for the major envelope glycoprotein
(gp90) and 8 for the transmembrane glycoprotein (gp45). Four MCAbs specific to epitopes of gp90 neutralized prototype EIAV infectivity.
These neutralizing MCAbs apparently reacted with variable regions of the envelope gp90, as evidenced by their unique reactivity with the panel of isolates, suggesting recognition of at least three different neutralization epitopes. The conformation of these epitopes appears to be continuous, as they resisted treatment with sodium dodecyl sulfate and reducing reagents. Monoclonal antibodies that reacted with conserved epitopes on gp90 or gp45 failed to neutralize EIAV.
Our data also demonstrated that there was a large spectrum of possible EIAV serotypes and confirmed that antigenic variation occurs with high frequency in EIAV. Moreover, the data showed that
x ii variation is a rapid and random process, as no pattern of variant evolution was evident by comparison of 13 isolates from parallel infections. These results represent the first production of neutralizing MCAbs specific for a lentivirus glycoprotein and document alterations in one or more neutralization epitopes of the major surface glycoprotein among sequential isolates of EIAV recovered during persistent infection.
Monoclonal antibodies were used to dissect the antigenic sites of the surface glycoproteins of the prototype cell-adapted Wyoming strain of EIAV. Serologic reactivities of these MCAbs were determined by ELISA, additive ELISA, competitive ELISA, and Western blot assays. The results indicated that antigenic reactivity of gp90 was localized on at least four distinct epitopes, two of which were important in neutralization. Our studies also revealed that these epitopes were localized on overlapping antigenic sites on gp90. On the other hand, only two distinct non-overlapping epitopes were identified on gp45. Competitive binding studies of neutralizing
MCAbs and reference EIA-positive horse serum delineated the presence of a neutralization domain on gp90 that appears to be immunodominant both in naturally infected horses and in mice immunized with EIAV.
Limited proteolytic fragmentation of the gp90 component of several serologically distinct EIAV isolates produced common 12K immunoreactive fragments that contained a conserved epitope. These results indicate the occurrence of conserved antigenic regions on
EIAV glycoproteins as well as a neutralization domain on gp90, which can be used as potential targets for vaccine development.
x iii CHAPTER I
INTRODUCTION
1.A. LITERATURE REVIEW
I.A.I. Equine infectious anemia: etiology and clinical disease
Equine infectious anemia virus causes a naturally occurring
disease in members of the family Equidae and is of worldwide
distribution (60). Equine infectious anemia virus is currently
considered to be a member of the subfamily Lentivirinae in the family
Retroviridae. Characteristics common to all retroviruses include the
nature of their genomes (RNA), their structure (similar overall
chemical composition), and their mode of replication through a DNA
intermediate. The last criterion, replication through a DNA
intermediate, distinguishes this RNA family of viruses from other RNA
viruses because any virus utilizing a virus-coded RNA-dependent DNA
polymerase (reverse transcriptase) for replication of its genome is
included in the family Retroviridae. The unique properties of EIAV are closer to those of visna, maedi and progressive pneumonia viruses and human immunodeficiency virus (HIV) than other retroviruses. The disease equine infectious anemia (EIA) is characterized by viral persistence, recurrent episodes of fever, hemolytic anemia, bone marrow depression, lymphoproliteration, and immune complex glomerulonephritis (52, 57, 60, 97). However, to date, no neoplasms appear to be caused by EIAV.
Transmission occurs naturally through the transfer of blood from the infected horse by blood feeding insects, particularly horse flies and deer flies. Experimental disease is readily induced by the
injection of infected blood or plasma by any parenteral route. The incubation period varies from five days to several months, and the
same inoculum given to a group of horses will often result in acute
deaths in some, while others develop chronic disease of episodic
nature during which there are minimal clinical manifestations (62).
Upon initial infection or after experimental inoculation, there is
rapid virus replication particularly in circulating and resident
macrophages and the virus can be found in high concentration in the
plasma. If horses die with acute EIA, lymphoid necrosis with lack of mature cells in the lymphoid germinal centers is a prominent finding.
The viremia with EIAV persists for the lifetime of the animal and
increases in virus titers are associated with the recurrence and
exacerbation of the acute illness, i.e., the chronic form of the
disease (95). Histological lesions in acute EIAV include
splenomegaly, lymphadenopathy, glomerulonephritis, and alteration in
the architecture of the liver due to focal degeneration, including
accumulation of lymphocytes and histiocytes (58, 73). Neutralizing
antibodies to EIAV are detected in horse sera years after the initial
infection (26), and most of thecirculating virus is found as
infectious complexes with antibody (96). Glomerulitis occurs in
horses with the active disease because of deposition of circulating
virus-anti body complexes (5). The erythrocytes of infected horses
are coated with complement and antibodies (67). The immunologically mediated active hemolysis most likely involves antiviral antibodies which are responsible for complement binding that results in
decreased erythrocyte life span, increased osmotic fragility, and erythrophagocytosis culminated by severe anemia (96). Equine
infectious anemia virus appears to replicate in vivo mainly in horse peripheral blood leukocytes or bone marrow which causes cellular
degeneration (64). However, it is interesting to note that although
viral antigens can be detected in virtually all tissues of infected
horses, macrophages are the predominant population of infected cells
(95).- The replication of EIAV in resident and circulating macrophages could account for the presence of the virus in almost all
tissues of infected horses (97). The agar gel immunodiffusion (AGID)
test has been the test of choice for the diagnosis of EIAV (27, 28,
105, 123, 128) by demonstrating the presence of precipitating
antibodies that react with the group specific antigen which were
demonstrated to be correlated with EIAV viremia (28).
I.A.2. Viral Characteristics
I.A.2.a. Physicochemical properties
Equine infectious anemia virus has a lipid-containing envelope with buoyant density of about 1.16 gm/cm (18, 92). It is sensitive
to lipid solvents but resistant to trypsin, RNAase, and DNAase. The
genome consists of RNA which is demonstrated by incorporation of
uridine but not thymidine. The replication is also sensitive to DNA
inhibitors in the early phases (52, 57). The RNA is single stranded with a molecular weight of about 5.5 X 10^ and sediments at 60S to
70S. The virion contains 4S RNA which is probably host derived tRNA,
and a small amount of 5S RNA of uncertain origin (18). The genome is
composed of two single-stranded subunits of approximately 34S with
C molecular weight of about 2.8 X 10 dalton.
I.A.2.b. Reverse transcriptase
The EIA virion contains a reverse transcriptase (RT) that can
catalyze synthesis of DNA from a synthetic RNA template or from the EIAV genome. The reverse transcriptase has a molecular weight of about 70,000 dalton and requires Mg++ as its divalent cation (2, 16,
17). Since EIAV contains a high molecular weight RNA genome, RT, from five to eight structural proteins, and matures by budding from the cell surface, it has been tentatively classified as a member of the family Retroviridae (16, 29, 60, 138).
I.A.2.C. Morphology
Equine infectious anemia virus displays a complex morphology consistent with that of most retroviruses. Electron microscopy studies confirmed that most retroviruses are roughly spherical structures of about 100 nm in diameter (151). A core or nucleoid is enclosed in an outer envelope made of a unit membrane, with spikes projecting from the outer surfaces. These spikes are particularly prominent in some retroviruses and are made of glycoproteins (127).
The size and shape of the EIA virion varies, but is generally spherical in nature ranging from 80 to 120 nm in diameter, with a rod- or cube-shaped electron-dense nucleoid of about 40 to 60 nm
(150, 151). It is bounded by a double-layered envelope with surface projections of about 6 to 8 nm in diameter (106, 139, 150) which are acquired from modified cellular plasma membrane during budding of the virus (93). Morphologically, EIAV is virtually indistinguishable from the lentivirus which causes acquired immune deficiency syndrome
(AIDS) in man (98).
I.A.2.d. Genomic and antigenic classification
Nucleic acid sequence homology was not detected between EIAV and members of the family Equidae, a variety of other mammals (126) and the following retroviruses: murine leukemia virus (MLF), mouse mammary tumor virus (MMTV), rat leukemia virus, the endogenous feline retrovirus (RD114), feline leukemia virus (FeLV), bovine leukemia virus (BLV), the endogenous hamster virus (HaLV), avian myeloblastosis virus (AMV), baboon retrovirus (BaEV), squirrel monkey retrovirus (SMRV), and Mason-Pfizer monkey virus (MPMV) (151). Thus,
EIAV is not thought to be an endogenous virus of the horse and is quite distinct from other retroviruses (151). Using broadly reactive anti sera to group specific antigen (gag) structural components, EIAV failed to react with a variety of oncoviruses (16). Morphologically,
EIAV resembles the lentiviruses (47); it also shares with them the trait of antigenic variation in the infected host (72, 102, 110,
129). Stowring et al. (138) previously reported no cross reactivity of EIAV group specific antigens with the lentiviruses visna, maedi, progressive pneumonia virus, and zwoegerziekte, using broadly reactive antisera. However, new significance arose by the inclusion of human immunodeficiency virus (HIV) in the lentivirus subfamily and therefore, lentiviruses became a rapidly expanding field of study.
A partial cross-reaction between the major internal protein of EIAV and p25 of lymphoadenopathy-associated virus (LAV) has been observed
(15, 98). Moreover, sequencing studies recently demonstrated that
EIAV gag and pol genes were found to be related to human
T-lymphotropic virus (HTLV) type III (HIV) and visna virus, which indicated that these viruses constituted a group of viruses clearly distinct from that of type C viruses or the bovine leukemia virus, and HTLV types I and II (137). Recently an interspecies radioimmunoassay (RIAs) was capable of demonstrating immunologic cross reactivity among the major structural proteins of EIAV, visna, and caprine arthritis encephalitis virus (CAEV)(153).
I.A.2.e. Structural proteins and glycoproteins
Equine infectious anemia virus possesses at least two
immunologically distinct components. One is associated with the
viral envelope and is glycoprotein in nature while the other is a
nonglycosylated protein localized inside the virion (17, 29, 59, 100,
112, 120). The former is demonstrated by neutralization and
haemagglutination tests and is considered to be type specific (69,
133). The latter is released from the particle by ether treatment
and is detected by complement fixation and immunodiffusion tests (70,
107), has a molecular weight of 26,000 (p26) and shows group-specific
activity.
Detailed cataloging of the viral polypeptides of the
cell-adapted Wyoming strain of EIAV was determined by high resolution
guanidine hydrochloride gel filtra tio n (GHCL-6F) and sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using
radioactive leucine or glucosamine labels for the cell-adapted
Wyoming strain of EIAV (100, 120). The chromatographic analysis of
EIAV revealed six distinct peaks of radioactivity. The first protein
eluted from GHCL-GF was contained in the void volume fractions of the
column with an apparent molecular weight (MW) of 100,000. The next
EIAV protein eluted from the gel has a MW of about 74,000. This
component was followed by the major EIAV structural polypeptide (p26) which displayed a MW of 26,000 and, in succession, by three
additional components of 15,000 (pl5), 11,000 (pll) and 9000 (p9) molecular weight respectively.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis of EIAV revealed four major low molecular-weight
polypeptides (p26, pl5, p ll, and p9) and six minor components. The minor components include two glycosylated polypeptides of 90,000
(gp90) and 45,000 (gp45) molecular weight and four minor
nonglycosylated proteins designated as p70, p61, p30, and p23.
Comparisons of the four major nonglycosylated proteins p26, pl5, pll,
and p9 by peptide mapping and enzyme-linked immunosorbent assays
demonstrated the unrelatedness of these four proteins (100). Further
characterization demonstrated that p26 and p9 focus at isoelectric
points (PI) 6.2 and 5.0 respectively, and they contain no unusual amino acids while pl5 displayed a heterogeneous isoelectric focusing pattern, with a PI ranging from 5.7 to 8.3, which was attributed to different levels of phosphorylated serine or threonine or both. The
pll of EIAV focused at a PI of >10, which indicates a high content of basic amino acids. The gp90 of EIAV appeared to be the more heavily glycosylated polypeptide, while the gp45 component aggregated in 6 M guanidine hydrochloride, reflecting a hydrophobic character. No disulfide linkages were detected between the EIAV glycoproteins
(120). Localization of the virion proteins was done by standard procedures in which purified virus was treated with the protease bromelain and the resulting subparticles isolated and analyzed by
SDS-PAGE. These studies demonstrated that gp90 and gp45 form the surface proteins of EIAV whereas p26, pl5, p ll, and p9 constitute the internal virion components with pll closely associated with the viral
RNA genome (100). 8
I.A.3. Virion surface glycoproteins
I.A.3.a. Neutralization
The first neutralizing antibodies in EIAV infected horses were
demonstrated by titratin g surviving virus in susceptible horses (136,
140). Kobayashi and Kono (64) found that EIAV could be propagated in
horse leukocyte culture, with resultant cytopathic effect (CPE).
Horse leukocytes cannot be routinely maintained for a long time and
endpoints of virus titrations were difficult to estimate. Horse
leukocyte cultures were used for virus neutralization (51, 67), and
antigenic differences between viral isolates were demonstrated
despite these technical difficulties (68). There was no
cross-neutralization between eight different isolates from several
different countries when studied by virus neturalization _in vivo
(71). This observation was supported by studies in horses where
cross-protection was lacking (73).
Neutralizing antibody against enveloped viruses is usually
directed at surface components (99). The virion surface of EIAV is
composed of a highly glycosylated protein of apparent MW of 90,000
dalton, which probably forms the knobs of the surface projections and
a glycosylated 45,000 dalton hydrophobic protein which may span the
lipid bilayer forming the spike of the projections (100, 120). This model is supported by data indicating that gp45 is released by
treatment of the virus with the protease bromelain as previously mentioned (100). In addition this protein was not susceptible to
lactoperoxidase catalyzed surface radioiodination of whole intact
virus, but was labeled after disruption of the virus with nonionic
detergent (20). 9
The contributions of both the carbohydrate and peptide portions of different retrovirus glycoproteins was assessed by glycosidase and
protease treatments on their reactivity with immune or hyperimmune
sera (8, 124, 131, 145). The results of these studies indicated that
the immunogenic portions of Friend murine leukemia virus gp71 (8) and
avian B77 leukosis virus gp85 (145) were exclusively protein in
nature, i.e., removal of the carbohydrates from these glycoproteins
had no significant effect on their reactivity. In contrast, the
reactivity of bovine leukemia virus gp51 (124, 131) and avian myeloblastosis virus gp85 (145) with homologous immune sera was completely abolished when the carbohydrate portion was stripped off
by glycosidase treatment. Recent studies with EIAV suggested that both carbohydrate and protein moieties of gp90 contribute to its
reactivity. The predominant reactivity, however, appeared to be against its peptide epitopes (103).
Mechanisms involving antibody-mediated virus neutralization
have been studied extensively (25, 31, 89, 90). Two general pathways of virus neutralization have been postulated. The first pathway
involves a physical reduction in the number of virus particles either by complement-mediated virolysis or by cross-linking of virus
particles with resultant virus aggregation. The second pathway
involves mainly the binding of antibody to viral antigens which
consequently inhibits one or more of the early steps (attachment, penetration, uncoating) necessary to establish a productive viral
infection. However, a novel pathway for neutralization of enveloped
viruses has recently been proposed (46). It has been shown that many enveloped viruses enter their target cells by endocytosis and are 10 subsequently located in cellular compartments of increasing acidity.
Fusion between the membranes of the virus particle and that of the prelysosomal endosome must take place before replication commences.
Gollins and Porterfield (46) showed that antiviral antibody can neutralize West Nile virus, a flavivirus, by inhibiting this intra-endosomal acid-catalysed fusion step.
The interaction of presumably “simple" viruses, e.g ., the naked picornaviruses, with antibody appears to involve only one specific antigen, as an immunogenic element (89, 90). This interaction is relatively straightforward and if neutralizing antibodies are involved, inactivation of the virus would be the outcome. However, with increasing complexities of viruses, the number of determinants increases and the interpretation of the effect of antibody on such viruses becomes more difficult. Critical, noncritical, and possibly quasi-critical antigenic sites have been proposed (89, 90). The interaction of these different antigenic determinants with antibodies of different specificities may not lead to inactivation of some viruses. For example, infectious immune complexes have been demonstrated in the blood of EIAV-infected horses (96). In view of the preceeding discussion on neutralization of viruses, diverse mechanisms could be hypothesized that might explain why immune complexes retain infectivity. The f ir s t mechanism might involve binding of antibodies (interfering antibody) to noncritical sites which in turn might block binding of potential neutralizing antibodies to critical sites. The second mechanism could involve saturation of possible antigenic determinants on EIAV or simple aggregation of EIAV particles that may not result in intrinsic 11 neutralization.
I.A.3.b. Use of monoclonal antibodies
The recent advances in somatic cellhybridization make it possible to produce virtually unlimited quantities of MCAbs of predefined specificity. These techniques, f ir s t developed by Kohler and Milstein (65, 66), are used extensively to produce hybridoma cjII lines that secrete MCAbs against viral antigens (44, 74, 91, 113,
147).
The strategy for immortalizing specific immunoglobulin-producing
B lymphocytes in culture is to fuse them with myeloma cells. The ensuing hybrids are then selected in the presence of hypoxanthine, aminopterin, and thymidine (HAT) medium. The aminopterin component of HAT medium blocks the de novo pathway of DNA synthesis, which is a prerequisite for cell replication. There is an alternative or salvage pathway by which normal cells are capable of utilizing hypoxanthine and thymidine as substrates to synthesize DNA. This pathway requires two key enzymes, hypoxanthine-guanine phosphoribose transferase (HGPRT) and thymidine kinase (TK). Myeloma cells used for fusion lack the TK and HGPRT and are selected for this property in the presence of either 5-bromo-2-deoxyuridine (BUdR) or
8-azaguanine. Unfused myeloma cells therefore cannot synthesize DNA through the alternative pathway in HAT selection medium. Although lymphocytes can u tilize both pathways of nucleotide synthesis, they cannot survive in vitro for a long period oftime due to their finite life span in culture. However, the hybrids between lymphocytes and myeloma cells will grow in HAT selection medium because of the availability of the key enzymes from the lymphocytes and the 12 contribution of immortality from myeloma cells.
Since MCAbs react with only a single antigenic determinant they
have proved to be remarkably specific biological probes for the analysis of protein polymorphism, as well as discerning the structural and functional properties of various viral proteins. To date, the emphasis of these studies were on the characterization and determination of viral proteins that serve as targets for virus-neutralizing antibodies.
Antigens displayed on the surface of virions or virus infected cells are the object of intensive investigations as they mediate infection of host cells and serve as primary targets for host immune responses (99). Monoclonal antibodies have been invaluable for the understanding of structural and functional features of these viral proteins. Epitopes recognized by specific MCAbs define structural and immunological domains of proteins. Functional maps of proteins can be inferred when antibody binding results in altered protein activity. In the area of vaccine development such functional maps can provide valuable clues in choosing synthetic peptide antigens to be tested as protective immunogens.
Immunity to viral infections in large part depends on the development of immune responses to antigens displayed on the surface of virions or virus infected cells. Identification of these important surface glycoproteins or proteins which elicit protective immunity, referred to as protective antigens, represents an important step in the development of effective viral vaccines. Polyclonal sera against purified glycoproteins of many viruses were found to neutralize virus infectivity. A good example in retroviruses is the 70,000 dalton glycoprotein (gp70) of murine leukemia virus (MLV).
This gp70 is the major surface component of this virus, and carries determinants against which neutralizing antibodies are directed
(53, 56, 63, 135). The envelope glycoproteins of MLV are believed to consist of two distinct domains (9), the amino terminal domain, which
is minimally conserved between MLV strains and is thought to be
involved in host cell receptor binding and type-specificity, and the carboxy terminal domain which is highly conserved and is believed to
serve a more structural role. Similarly, an important determinant of virus neutralization on felineleukemia virus (FeLV) was also
localized on the major surface glycoprotein of 70,000 dalton.
Interestingly the MCAbs used to define this determinant was found to be broadly reactive, i.e., most FeLV isolates tested from FeLV subgroups A, B, and C, were neutralized (114).
Protective surface antigens have been identified for many viruses and often more than one surface glycoprotein is important in
immunity to such viruses. The hemagglutinin (HA) of orthromyxoviruses is required for attachment of virus to host-cell receptors and for fusion with intracellular membranes. This surface antigen stimulates neutralizing antibodies, while neuraminidase (NA), the other surface glycoprotein, is required for release of virus from infected cells, and it stimulates antibodies that impede release of virus from infected cells and hence reduce the quantity of virus produced during infection (3, 75, 77). Monoclonal antibodies directed against the surface HA and NA of influenza virus protect mice against challenge with virulent wild-type virus, whereas other monoclonals directed against internal nucleoproteins or matrix 14
proteins do not alter the course of the disease (3).
It should be noted, however, that immune responses to certain
internal viral proteins may play a cooperative role in the
development of effective immunity. A large body of information is
accumulating on the interaction of these internal viral proteins and
cell membranes. Lamb et al. (80) studied the M 2 protein of influenza
A virus. This protein is translated from a spliced mRNA sequence
transcribed from genome segment 7, and is a 97 amino acid
non-glycosylated protein with a single hydrophobic region between
residues 25 and 43. Lamb et al. (80) showed that is present in
the infected cell and is inserted into the plasma membrane, with its
N-terminus external to the cell. They suggested that this protein
could be an important target in natural infection since they were
able to demonstrate that anti-peptide antibodies could recognize the external portion of this protein. In particular, they speculated
that M2 could be a major viral protein displayed on infected cell membranes which are recognized by cytotoxic T-cells.
The major protective antigens of naked viruses are also located on their surfaces. One example would be foot and mouth disease virus
(FMDV) which consists of 8kb single stranded RNA surrounded by 60 copies of four structural polypeptides (78). Only VP1 has been associated with ability to induce virus neturalizing antibodies (4).
However, recent reports on polio virus indicated that epitopes
located on VP2 and VP3 in addition to those clustered on VP1 can induce neutralizing antibodies (32). Studies with polyclonal and
MCAbs to poliovirus type 1 capsid protein VP1 confirmed that the virus underwent conformational changes during neutralization (36, 15
89). Poliovirus assumes two isoelectric forms with pis of about 7
and 4, of which only the former is infectious. When virus is
neutralized, the pi is shifted and frozen at 4. The neutralization
and pi shift occur only if antibody molecules bind bivalently to the
virion. An average of four bound antibodies per virion are required
to neutralize the virus (55).
Monoclonal antibodies have proved to be very important reagents
to discern the functional and structural properties of viral
proteins. The number and location of antibody binding sites on the HA
and NA of influenza viruses were determined using MCAbs in selecting
variants and in competitive binding assays which demonstrated the
presence of three or four nonoverlapping antibody binding sites (10,
144, 148).
The specificity of the reaction of MCAbs provide the basis for
the detection of subtle differences between the gene products of
related viruses and thus the differentiation of one virus from
another. Indeed this approach greatly exceeds conventional serology
in specifity, and has lead to the recognition of differences between
viruses previously thought identical. Human convalescent serum for example, failed to detect serological differences between human
respiratory syncytial virus (RSV) isolates. However, MCAbs were
capable of distinguishing between different human RSV isolates and
therefore different types of human RSV were suggested to exist (45).
Similarly, a wide antigenic variation was suggested to occur
naturally among field strains of coxsackie B virus type 4 (CBV-4).
Cao et al. (14) were able to identify five different antigenic
variants among nine virus strains by using a panel of nine neutralizing MCAbs to CBV-4. Comparing nine transmissible gastroenteritis virus (TGEV) strains by using a panel of MCAbs, Laude et al. (81) confirmed the existence of a close antigenic relationship between these strains but revealed also the occurrence of distinct antigenic differences at the same time. Thus, the situation with
TGEV seems to differ from the murine coronavirus (MHV-4) where each isolate was reported to have a unique pattern of reactivity to a set of anti-glycoprotein MCAbs (39, 81). The laboratory strains of rabies virus were found to be different from each other as well as from rabies-related viruses (38). These differences were attributed to different antigenic determinants on the glycoproteins of these viruses.
The discriminatory power of MCAbs make them ideal reagents for investigation of antigenic relatedness between viral proteins. They are invaluable reagents in demonstrating minor antigenic differences between closely related proteins. This is well illustrated by recent work with rabies and polio viruses, where antigenic differences between closely related proteins could be delineated between respective field and vaccine strains previously considered to be antigenically indistinguishable (38, 87). This finding is of particular importance and may explain why vaccines occasionally fail to protect against infection by field strains. The appearance of antigenically variant poliovirus type 3 strains in a vaccinated population in Finland document these facts (87). In fact, the polio viruses isolated during the outbreak were antigenically heterogeneous when examined for neutralization by a large panel of MCAbs. Most of the strains were unusual antigenically in that they were not 17 neutralized by the majority of the antibodies tested and the antigenic properties were consistent with the variation found in the sequence of the immunodominant antigenic site involved in the neutralization of the virus. It is, however, not clear why the heterogeneity arose, as this site is generally well conserved among unrelated type 3 polio viruses (87).
I.A.3.C. Antigenic variation
The survival and circulation of many pathogens in nature despite the induction of a rigorous immune response in the infected host depend on the inherent capabilities of these pathogens to change the antigenic determinants of their surface proteins during the course of an infection. Antigenic variation of various pathogens, including parasites, bacteria and viruses, is now well documented and partially understood at the molecular level. This property has been an important factor contributing to failures in the control of these pathogens by conventional vaccination. A prime example of antigenic variation has been observed in African trypanosomes which evade the immune responses of their mammalian host by sequentially expressing a series of variable surface glycoproteins (VSGs) from a library of
VSGs genes (34). This enables certain populations of trypanosomes of an infected animal to keep "one step ahead" of the antibodies raised against their VSGs. Under most conditions each trypanosome expresses one, and only one, VSG on its surface at a given instant (11, 30).
Antigenic variation represents a potent survival strategy whereby viruses can eludethe antiviral immune response by emergence of antigenically new virus strains. These are due to mutational changes of the viral genome. Once these changes are manifested in the antigenicity of the virus, the virus can persist and circulate in the
population rather freely. An excellent example of antigenic
variation is in type A human influenza viruses (119, 149). The
variability of influenza viruses appears to involve several mechanisms, of which reassortment, point mutation, addition or
deletion in genes coding for both surface glycoproteins, HA and NA,
have been described. Of these mechanisms, reassortment in genes
coding for both surface glycoproteins induces major alterations in
the virus antigenicity, which is recognized as "antigenic shift." On
the other hand, minor mutations in the genome, leading to one or more
amino acid sequence changes may alter the antigenic sites in such a way that they are no longer recognized by the host's immune system,
i.e., " antigenic drift" (149). However, in an infected host,
alteration of the antigenic structure of the infecting virus allows
the antigenic variant to escape from neutralization by antibodies,
thus perpetuating the infection in an otherwise immunologically
competent host.
Antigenic variations were also demonstrated in individual
animals infected with the lentivirus, visna. However, viruses
recovered from sheep peripheral blood leukocytes prior to and several months after development of antibodies were antigenically identical
to the parental strain used for inoculation. Longitudinal studies of
visna viruses collected from two infected sheep were interesting in
that they showed that more than one strain of virus could co-exist in
the animal (108, 109, 110, 111). The selection of antigenic variants
under antibody pressure was substantiated by experiments done in
vitro using sheep cell cultures that were inoculated with 19
plaque-purified visna virus and maintained with antibody in the
medium (111). This antigenic variation of visna virus has been
correlated with point mutations clustered in the 3' terminal region
of the viral RNA (22).The same workers further found that these
changes were clustered in a single region of the viral genomes of
seven antigenic variants and suggested that these genomic changes are
linked to the antigenic phenotype of these variants (23).
Equine infectious anemia is a unique retrovirus infection in
that the clinical disease is characterized by bursts of viremia which
occur in sequential episodes separated by several weeks or months
(29, 60, 68). Kono (68) and Kono et a l . (69, 72) have suggested that
antigenic variation is likely to be the mechanism operable in EIAV to
explain viral persistence and the recurring episodes of clinical
illness in EIAV. These workers reported that sera taken at various
times from an infected animal were able to neutralize virus isolates
from previous clinical episodes, but did not effectively neutralize
subsequent virus isolates. They further confirmed the antigenic
drift of EIAV in infected horses by using antibodies raised in
rabbits against these viruses (69, 72). Recently, alterations were observed in the RNA genome of several EIAV isolates by using oligonucleotide fingerprinting technique (121). At the same time, structural alterations of the surface glycoproteins of EIAV were demonstrated during persistent infections (102, 129). They were evident in peptide and glycopeptide maps with several peptide additions and/or deletions occurring between each isolate and its predecessor. These genotypic and phenotypic changes demonstrate that
EIAV is a highly mutable virus and suggest that point mutations occur 20
frequently in genes coding for viral glycoproteins (102, 121,
129). It is probable that the emergence of novel antigenic strains of EIAV is a manifestation of these alterations.
Genomic heterogeneity has also been firmly established as a
prominent characteristic of the AIDS virus, HIV (6, 152). In
addition, the envelope gene of this virus is relatively hypervariable
in comparison with the remainder of the viral genome. Recently, Hann et al. (49) reported that AIDS viruses isolated sequentialy from persistently infected individuals appear to have evolved in parallel
from a common progenitor virus, as examined by Southern blot genomic analysis, molecular cloning, and nucleotide sequencing. The same workers further confirmed that virus isolates from any one patient were all much more related to each other than to viruses from other
individuals. The variation in the envelope proteins of EIAV and visna as well as the genetic variation in HIV may give rise to viruses with altered antigenicity/immunogenicity, virulence, tissue tropism, or drug sensitivity. These in turn may affect viral pathogenesis and allow the virus to escape the host's immune defenses.
I.A.4. Viral infected cell membrane antigens
Many proteins destined to become integrated into cellular membrane (including all the viral glycoproteins) are synthesized on membrane bound polyribosomes (76). All these proteins are synthesized with an amino-terminal peptide extension termed a "signal peptide" or "leader" sequence. These viral glycoproteins are attached to the lipid bilayer by a single transmembrane anchoring peptide, the bulk of their structure being on the external side of the membrane with a small domain on the cytoplasmic side. Cells 21
infected with retroviruses express a number of proteins on the cell
surface, e.g ., gp70 of murine leukemia virus (24).
Specific EIAV antigen was demonstrated on the surface of
persistently infected fibroblasts using both a radioimmune binding
test (94) and lymphocyte cytotoxicity. Antibody against this antigen
appeared within one month after infection and persisted at least four years. However, j_n vitro killing of cells infected with EIAV by
antibody-dependent cell-mediated cytotoxicity has been difficult to
demonstrate (41). This was attributed to low antigen density
expressed on the cell surface. In addition, low level jji vitro
killing of EIAV infected cells by lymphocytes from infected horses was also demonstrated (41). The antigen type expressed on EIAV
infected cells is not known; however, the presence of these surface
antigens provides several potential mechanisms for destruction of
infected cells in vivo.
I -B- RESEARCH OBJECTIVES
The main thrust of the present research is to get a better
understanding of EIAV antigenic variation as a mechanism for viral
persistence and to define determinant(s) capable of stimulating neutralizing antibody. The specific objectives are listed below along with their relative significance.
1) To generate a diverse panel of monoclonal antibodies to the surface glycoproteins (gp90 and gp45) of the prototype strain of EIAV. These monoclonal antibodies will be used to identify and dissect the antigenic sites of the EIAV glycoproteins.
2) To define the pattern and extent of serological cross-reactions of EIAV isolates. The serological reactivities of 22
the EIAV isolates will be determined with MCAbs in Western blot and
ELISA. These studies will allow us to compare the glycoproteins
(gp90 and gp45) of different EIAV isolates with MCAbs.
3) To study antigenic variation of EIAV and delineate the changes in the virion surface glycoproteins. Using a diverse panel
of MCAbs, it is possible to dissect or map out the altered
antigenicity of the surface glycoproteins present in isolates from
ponies with chronic EIA. A detailed serological comparison of viral glycoproteins of different EIAV isolates recovered from animals
infected in parallel with the same inoculum might help in defining
the extent and pattern of the evolution of EIAV variants. It may
help determine the number of EIAV serotypes that may exist and will assist in cataloging EIAV into serological groups.
4) To determine which virus component(s) are involved in
the antigenic changes observed in EIAV variants. Using the diverse
panel of MCAbs we will see if antigenic variation is limited to gp90 or gp45 or whether both components contribute significantly to the observed altered antigenicity of EIAV variants.
5) To define the determinant(s) involved in stimulating
neutralizing antibodies to EIAV. The identification of unique and/or conserved epitope(s) which are necessary targets for neutralizing antibodies may represent an essential first step for designing a
synthetic vaccine. If conserved or group-specific epitopes can be
identified which possess protective immunogenicity, then the generation of an effective group-specific immunogen would be feasible. CHAPTER I I
GENERATION OF MONOCLONAL ANTIBODIES: PRODUCTION AND CHARACTERIZATION
II.A. INTRODUCTION
Equine infectious anemia virus, a retrovirus in the subfamily
Lentivirinae, causes a naturally occurring disease in all members of
the horse family. The disease has a worldwide distribution and is of great economic importance to the horse industry. The unique
feature of EIA is the persistent nature of the infection which
involves bursts of viremia associated with recurring clinical
symptoms, i.e., the chronic form of the disease (29, 60, 72). Within
the lentiviruses, EIAV is closely related to, and shares additional
features with the human immunodeficiency virus (98, 137).
Equine infectious anemia virus has been found to contain two main surface glycoproteins (gp90 and gp45) and four major
nonglycosylated internal proteins designated p26, pl5, pll, and p9
(100, 120). The major glycoproteins of the reference Wyoming strain of EIAV were found to have different electrophoretic mobilities from
those of antigenic variants, while the major core proteins were
identical. Research to date has shown that EIAV variants contain
structural alterations confined to the envelope glycoproteins and that these changes are paralleled by changes in viral RNA (102, 121,
129). If these genotypic and phenotypic changes are mirrored antigenically, EIAV will possess an essential mechanism for persistence and spread.
A more comprehensive understanding of the molecular basis of the altered antigenicity and immunogenicity is needed in order to produce a vaccine for EIA. To obtain knowledge about the antigenic
23 2H alterations, MCAbs were produced against antigenic determinants on the surface glycoproteins of EIAV. Serological reactivities of a panel of MCAbs were monitored in enzyme-linked immunosorbent assay,
Western blot immunoassay and neutralization assays. Since results have shown that sequential isolates obtained from persistently infected ponies differ in their surface proteins, our aim was to use these glycoprotein specific MCAbs to identify and dissect the antigenic sites on EIAV isolates obtained from persistently infected ponies. This will help in mapping out the altered antigenicity of the surface glycoproteins which may be of relevance to the observed recurrent clinical disease.
II.B MATERIALS AND METHODS
II.B .l. Cell Culture
Fetal equine kidney (FEK) cells and fetal donkey dermal (FDD) cells used for virus production and neutralization assay were prepared in the laboratory, maintained in Eagle's minimal essential medium (GIBCO Laboratories, Grand Island, NY), and supplemented with
5 to 10% sterile fetal calf serum (GIBCO), 25 mM HEPES
(N-2-hydroxyethylpiperazine-N1-2-ethanesulfonic acid) buffer, 10 mM sodium bicarbonate, 150 U of penicillin G per ml, and 150 ug of streptomycin sulfate per ml. Persistently infected cultures were p maintained in roller bottle cultures (850 cm growth area) containing
50 ml of medium.
II.B .2. Virus propagation and purification
The cell-adapted Wyoming strain of EIAV (88) was propagated in primary FEK cell cultures. The procedure for virus purification has been described in detail (100). Briefly, the virus was propagated in primary FEK cells which remain persistently infected. The cultures
were maintained in roller bottles and supernatant fluid was harvested
every 3 to 4 days, clarified by centrifugation at 10,000 xg for 30
min in a Sorvall GSA rotor, and concentrated about 10 to 40 fold by
using a pell icon membrane f ilte r (nominal exclusion, 10® dalton molecular weight, Millipore Corp., Bedford, Mass). The concentrate was centrifuged through an underlayered 10 ml cushion of 10% sucrose
and pelleted by centrifugation in a type 19 rotor (Beckman
Instruments, Inc., Fullerton, Calif.) at 19,000 rpm (50,000 xg) for 2
h. Pelleted virus was resuspended in 0.01 M phosphate buffer (pH
7.2) and sedimented to its equilibrium density (1.18 g/ml) on a 3 ml
20 to 80% (V0L/V0L) glycerol gradient (100, 102, 120) in a Beckman
SW-28 rotor centrifuged at 28,000 rpm (130,000 xg) for 4 h. The
virus band was collected, diluted in sodium phosphate buffer to a
final glycerol concentration of approximately 10% and pelleted by centrifugation in a Beckman SW-28 rotor at 28,000 rpm (130,000 xg)
for 90 min. The final virus pellet was suspended in 0.01 M sodium
phosphate buffer and stored at -70°C in small aliquots. All
procedures were carried out at 4°C.
II.B.3. Glycoprotein purification
The glycoproteins of EIAV were purified by affinity chromatography over a column of lentil lectin bound to Sepharose in the presence of detergent as described previously (101, 102).
Briefly, 20 mg of purified EIAV was treated with 10 volumes of cold acetone to remove excess lipids and to precipitate the proteins of the virus. The proteins were collected by centrifugation at 10,000 xg for 10 min at 4°C and the pellet was solubilized by incubation for 26
15 min in buffer A (Tris [0.02 M], Nacl [0.1 M], pH 8.3) containing
0.5% sodium deoxycholate (Sigma Chemical Co., St. Louis, MO).
Insoluble material was removed by centrifugation at 10,000 xg for 10
min. Solubilized proteins were then separated by affinity
chromatography over a column of lentil lectin-Sepharose (Pharmacia
Chemicals, Piscataway, NJ) pre-equilibrated with buffer A containing
0.1% sodium deoxycholate. The sample was allowed to adsorb for 4 h
at room temperature; unbound material was washed through with the
same buffer. Bound glycoproteins were eluted with 0.2 M
methylglucoside in buffer A containing 0.1% sodium deoxycholate and
dialyzed for 36 h against 0.1 M ammonium bicarbonate, pH 8.0 and 12 h
against 0.05 M ammonium bicarbonate, pH 8.0. The proteins were then
lyophilized to dryness, resuspended in 0.01 M phosphate buffer, pH
7.2, and the protein content was analyzed by the Lowry technique and
polyacrylamide gel electrophoresis (85, 100).
II.B .4. Immunization of Mice
Two month old female BALB/c mice were injected subcutaneously
(sc) and intraperitoneally (ip) with either 100 yg of purified
glycoproteins of EIAV or 200 yg of intact virus or deoxycholate
disrupted virus in Freund's complete adjuvant (FCA). The mice were
injected on day 14 and 28 with the antigen perparations (same concentration and same route as above) in Freund's incomplete adjuvant. The final injection was given on day 88 with antigen in
0.01 M phosphate buffer, either intraperitoneally or intravenously
(iv) 4 days prior to fusion. Four types of antigen preparation were
used as the immunizing antigen: 1.) lentil lectin purified glycoproteins, 2.) intact whole virus, 3.) intact whole virus given 27
iv, and 4.) EIAV dissociated with deoxycholate.
II.B.5. Preparation of cells
II.B.5.a. Myeloma cells SP2/0-Agl4
This special cell line is characterized as a non-immunoglobulin
secreting myeloma cell. These cells lack the enzyme
hypoxanthine-guanine phosphoribose transferase (HGPRT) which makes
them unable to synthesize purines by the salvage pathway. They are
susceptible to the folic acid antagonist aminopterin which blocks de
novo biosynthesis of purines and pyrimidines. Aminopterin blocks the
dihydrofolate reductase enzyme and leaves the cell unable to produce
purines via de novo pathway synthesis. This classical biochemical
selection procedure, introduced by Littlefield (84), is widely
applied and is based on the use of either HGPRT-negative or thymidine
kinase-negative tumor cells. Thus, an HGPRT-deficient cell in media
containing aminopterin cannot synthesize or salvage purines and
rapidly dies. The use of 8-azaguanine resistant myeloma lines allows
the selective destruction of unfused myeloma cells and myeloma cell
to myeloma cell fusion products. The spleen cells which have
hybridized to myeloma cells can, however, survive in the presence of
aminopterin due to the presence of HGPRT activity from the spleen
cells. In addition, SP2/0-Agl4 (SP2/0), is a total nonproducer
variant from a hybridoma involving fusion of MOPC-21 and BALB/c
spleen cells. As a routine practice in our laboratory, these cells were passaged in medium containing 8-azaguanine (20 yg/ml) to prevent
them from reverting back to production of the HGPRT enzyme. After
passage, the cells were maintained in growth medium free of
8-azaguanine. 28
SF2/0 cells were seeded 3 to 4 days prior to anticipated use in 5 5 a fusion usually at a range of 1 X 10 to 7 X 10 per ml. The day
before the fusion, immunized mice were checked for antibody
production against EIAV antigens using an ELISA and those mice with
highest antibody titers were selected for spleen donors.
II.B.5.b. Feeder layer:
Thymocytes were found to be extremely useful as a feeder layer, which may provide undefined factors for continuous growth of hybridoma lines. The use of a feeder layer is particularly critical during cloning procedures (83, 115). Mice were killed and their thymuses were collected, minced and washed 3X in calcium and magnesium free Hanks' balanced salt solution (CMF Hanks' BSS) containing 100 U of penicillin 6, and 100 yg of streptomycin sulfate per ml. The thymocytes were then suspended at 3 X 10® to 5 X 10® cells per ml and 0.1 ml added to each well where appropriate.
II.B.5.C. Splenic lymphocytes
Spleens of immunized mice were collected aseptically in 50 ml centrifuge tubes. All procedures beyond this point were carried out in a laminar flow hood. Each spleen was washed with several changes of CMF Hanks' BSS and then transferred to a small petri dish, rinsed, and excess omentum removed. The spleen was then transferred to a second petri dish and reduced to a single cell suspension via mincing by scissors and the "blow out" technique. The "blow-out" technique involves the use of a 3 ml plastic syringe filled with CMF Hanks' BSS and fitted with a 25-gauge needle. The needle is inserted into the spleen capsule and the medium slowly injected. Multiple perforations and simultaneous injections of CMF Hanks' BSS help to express most of 29 the lymphocytes out of the spleen. The resulting cell suspension was transferred to a centrifuge tube and allowed to stand briefly to allow large debris to settle. After this, the cell suspension was decanted carefully to a second centrifuge tube, washed twice in CMF
Hanks' BSS (200 xg for 5 min), and resuspended in serum-free RPMI
(Roswell Park Memorial Institute 1640) medium. Total cell count and viability (trypan blue exclusion) were done simultaneously using a modified Neubauer hemacytometer chamber.
II.B.6. Hybridization
Viability of both lymphocyte and myeloma cells preparations used for the fusion was more than 90%. The spleen cells and myeloma cells
(SP2/0) were washed at least twice in CMF Hanks' BSS and resuspended in RPMI, and then mixed at a ratio of 1 SP2/0 cell to 4 spleen cells.
The resulting mixture was washed and pelleted for 5 min at 200 xg in a 50 ml centrifuge tube. Most of the supernatant medium was removed by aspiration, the pellet was broken with gentle external mixing and the tube was placed in a 37°C water bath.
The fusion protocol was derived and modified from that of Galfre and Milstein (42) and Shulman et al. (134). Basically polyethylene glycol (PEG) was added to the cell mixture dropwise with gentle stirring over a period of 1 min. The mixture was further stirred gently for another min using the same pipette. To gradually dilute the PEG, 2 ml of growth medium were added dropwise while stirring over a period of 2 min. An additional 3 ml of growth medium followed by 5 ml of growth medium were added in the same manner as described above over a period of 1 min each. Using the above described protocol O we were able to fuse as many as 4.0 X 10 spleen cells/tube using 1 30
O ml of PEG for every 1.6 X10 spleen cells and 10 ml of growth medium to dilute the PEG. The suspension was centrifuged at 200 xg for 5 min and the media removed by aspiration. The remaining cell pellet was gently broken, the cells were resuspended at 7.3 X 106 spleen cells per ml in growth medium and dispensed from a 5 ml pipette at 2 drops per well of a 96 well flat-bottomed m icrotiter plate. 5 Thymocytes were added as a feeder layer at 3 X 10 cells per well in
100 yl aliquot.
II.B .7. Maintenance and specificity testing of hybridomas
II.B.7.a. Care and selection of hybridomas
Cultures were briefly inspected daily under the microscope to monitor their progress. The fusion of cells by PEG is essentially a random and inefficient process, thus, cellular hybridization occurs in a low percentage of cells. Consequently, the selection of hybrid cells from residual parental cells was initiated 1 day post fusion by the addition of 100 yl growth medium containing hypoxanthine, aminopterin, and thymidine (HAT) daily for four days post fusion and by removing one half of the spent medium and replacing it with fresh medium. The media was allowed to trickle down the side of the well so that the cells were not dispersed and clones were countable.
Cells were fed medium containing hypoxanthine and thymidine (HT) on day 5, 6, and 7 post fusion. Screening for antibody production using enzyme-linked immunosorbent assay (see below) started when 10% of the wells turned acidic or when colonies attained a size of 1 to 2 mm in diameter. Cells generally reached this saturation density at 10 to
12 days post fusion. Hybn'doma clones of this size usually produce immunoglobulins at detectable levels and the percentage of false 31 negative results decreases.
II.B.7.b. Cloning by limiting dilution
After a hybridoma secreting a useful antibody was detected, cells from the appropriate well were immediately cloned. This is a necessary step for the following reasons. First, the hybridoma may be unstable and a variant, non-immunoglobulin secreting hybridoma may overgrow the hybridoma producing useful antibodies. There may also be more than one clone of hybridoma in the original well, and cloning will separate the desired hybridoma line from irrelevent cells.
Cloning by limiting dilution in liquid medium is a well accepted method (115). After a positive parent well was selected, it was cloned by limiting dilution in which limited numbers of cells were added to 96 well microtiter plates. Theoretically, the number of cells in the original well was counted and diluted appropriately so that every 2 rows of a 96 well microtiter plate contain 4 hybridoma cells/well, 2 cells/well, 1 cell/well and 50% of the wells in the last 2 rows contain 1 cell each.
The number of colonies per well was determined at approximately day 7 to 10 and single clone wells were selected for testing. Single positive clones were recloned a second time. The single clones from the second limiting dilution were assayed for antibody production and if all the clones were positive, the clone was considered to be stable. Cells were then expanded from 96-well plates to 24 well tissue culture clusters and finally transferred to flasks for ultimate cryopreservation and ascitic fluid production.
II.B.7.C. Cryopreservation and thawing of clones
Hybridoma cells were grown to high densities, resuspended at 5 X 32 c 10 cell per ml in cold growth medium, supplemented with 20% fetal
calf serum and 10% dimethyl sulfoxide (DMSO), and placed in glass
ampules. The ampules were sealed and placed immediately into a
cooling chamber where the rate of cooling was approximately 1°C per
min. Once frozen to -80°C, they were transferred directly into
liquid nitrogen.
Careful handling of the hybridoma cells during thawing is very
critical to assure good v iability. The ampules were retrieved from
the liquid nitrogen tank and immediately placed in a 37°C water bath
where they were rapidly thawed. When thawed, the ampule was placed
in a laminar flow hood, swabbed with 70% alcohol and the contents
were transferred to a 15 ml centrifugation tube. Cold HT medium (10 ml) was added slowly to the cell suspension over about 2 min to
gradually dilute the cells and preservative. This was necessary
because sudden dilution in the preservative DMSO has been reported to
cause severe osmotic damage to the cells. The cells were then
pelleted by centrifugation at 200 xg for 5 min, and resuspended in
fresh HT medium for culture.
II.B.8. Preparation of ascitic fluid
Hybridoma cells grown in the peritoneal cavity produce as much
as 20 mg of immunoglobulin per milliliter of ascitic fluid (113).
Therefore BALB/c mice(preferably male 12 to 20 weeks old) were
primed intraperitoneally with 0.5 ml pristane (2, 6, 10,
14-tetramethylpentadecance, Aldrich Chemical Company, Milwaukee, WI).
This is very important as the success rate of tumor development and
the probability of ascites formation is highly increased when mice
are preinjected with pristane (12). One to four weeks after priming 33 with pristane, mice were injected intraperitoneally with cloned hybridoma cells (1 X 10^ to 2 X 10^ cells per mouse). At 10 to 14 days after injection, the ascitic fluid was aseptically collected and clarified at 1000 xg for 10 min, and stored at -20°C.
II.B .9. Determination of immunoglobulin class
Fluids from cultured hybridomas usually contain small amounts of antibody and it was necessary to concentrate the fluids by ammonium sulfate precipitation (50% saturation) to detect immunoglobulins in immunodiffusion tests. The immunoglobulin isotype was determined by immunodiffusion using commercially available antisera: goat anti-mouse IgGl, IgG2a, IgG2b, IgM, IgA (MeloyLaboratories) and rabbit anti-mouse IgG3 (Miles Laboratories, Inc.). Immunodiffusion tests were performed in 1% Noble agar plates (borate buffer 0.15 M, pH 8.4). About 40 pi of concentrated supernatant were arranged radially in individual peripheral wells and 40 pi of 1:4 dilution of anti sera were placed in the central well. Plates were incubated overnight and precipitin lines were developed between the monoclonal immunoglobulin and the appropriate isotype-specific antibody.
II.B.10. Screening of hybridomas
To decrease the possibility of false negatives, hybridomas were screened when the cells reached maximum density or the clone size ranged from 1 to 2 mm. This usuallyoccurred from 10 to 28 days after fusion. A sensitive, simple and rapid enzyme-linked immunosorbent assay (ELISA) was developed so that an immediate decision about cloning the hybridoma could easily be reached. A solid phase enzyme irranunoassay was modified from that described by
Voller et al. (146) and O'Sullivan et al. (118). Purified virions were disrupted with 0.5% deoxycholate in Tris-saline buffer
(0.02 M/0.1 M) and maintained at 200 yg/ml concentration. They were diluted in coating buffer TEN (Tris [0.05 M], EDTA [0.001 M], NaCl
[0.15 M]) to a concentration of 20 yg protein/ml and dispensed at 1 yg/well and allowed to adsorb onto each well of a 96-well f la t bottom m icrotiter plate (Dynatech Immulon 1) for 16 to 20 h at 37°C incubation. The antigen was fixed with paraformaldehyde (4% Wt/Vol9 pH 7.2) for 5 min and nonadsorbed antigen was washed away with TEN buffer. The wells were blocked for nonspecific binding by incubation with 5% bovine serum albumin in phosphate-buffered saline (PBS) for 2 h at room temperature. The tissue culture supernatant or ascitic fluid to be tested for antibody was added (50 yl/well) and allowed to react for 1 h at room temperature. The plates were washed 3 times with TEN and 50 yl of peroxidase-labeled goat anti-mouse immunoglobulins (1:1000 dilution in TEN, Cappel, Cochranville, PA) was added and incubated for 45 to 60 min at room temperature. The plates were washed as before and developed using the substrate ABTS
(2, 2'-azino-bis(3-ethylbenz-thiazoline-6-sulfonic acid), diammonium salt (Sigma). The reaction was continued for 10 to 20 min and the optical density at 490 nm was recorded with a spectrophotometer
(Dynatech Lab. VA).
Positive hybridomas were also tested for ability to bind protein
A. The same procedure was used as above except that protein A peroxidase conjugate was used at a 1:500 dilution, and the conjugate and washing buffers contained 0.1% Tween 20. 35
II.B.ll. Identification of viral proteins by the protein blot
immunoassay "Western blot"
Unlabeled purified EIAV proteins were separated by sodium
dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and
the separated proteins were electrophoretically transferred to
nitrocellulose membranes (13, 143). The nitrocellulose membrane was
then saturated with 3% bovine serum albumin (BSA) in Tris buffer
saline (TBS) for 30 min, incubated with the MCAb or serum from
immunized mice for 90 min at room temperature, and washed in at least
three changes of Tris-saline containing 0.05% Nonidet P-40 (NP-40,
Particle Data Laboratories, Elmhurst, IL). The nitrocellulose membrane was then incubated for 1 h in Tris-saline containing a
1:1000 dilution of peroxidase conjugated goat anti-mouse
immunoglobulins. After incubation with the second antibody, the membrane was washed as above and incubated with the color development
solution. This solution was prepared by dissolving 60 mg of
4-chloro-l-naphthol (Sigma, Chemical Co. St. Louis, M0) in 20 ml of
ice-cold methanol and then mixing i t with 100 ml Tris-saline plus 60
yl 30% (Wt/Vol) hydrogen peroxide. The immunoreactive proteins
stained purple with this reagent.
II.B.12. Preparation of reagents
II.B .12.a. Growth media
Growth media (RPMI 1640, GIBC0 Laboratories, Grand Island, NY) was prepared from powdered medium supplemented with 24 mM sodium
bicarbonate, 15% fetal calf serum, 1% 200 mM L-glutamine, 1%
non-essential amino acids, 50 U of penicillin G per ml, and 50 yg of
streptomycin sulfate per ml. 36
II.B.12.b. Polyethylene glycol (PEG)
The PEG (MW 1450, Kodak Laboratory Chemicals, Rochester, NY) was
autoclaved at 15 PSI, for 15 min and kept in a 56°C water bath. One
ml of RPMI was added per gram PEG. The pH was adjusted to 7.2 under
sterile conditions with 0.1 N sodium hydroxide or 0.1 N hydrochloric
acid as required. The PEG medium was dispensed in small quantities
and stored at -20°C.
II.B.12.C. Hypoxanthine-Thymidine (HT)
A 100 X stock solution (hypoxanthine 1.0 X 10"^ M, thymidine 1.6 -3 X 10" M) was prepared by dissolving 136.1 mg of hypoxanthine (Sigma,
St. Louis, M0), and 38.75 mg of thymidine (Sigma) in 100 ml of
d istilled water warmed to 60-70°C. The HT medium was f ilte r
sterilized (0.45 pm) and stored in aliquots at -20°C.
II.B.12.d. Ami nopterin
A 100 X stock solution (4.0 X 10”5 M) was prepared by suspending
I.76 mg of ami nopterin (Sigma) in 50 ml of distilled water,
dissolving it by the dropwise addition of 0.1 N sodium hydroxide, and
adjusting the volume to 100 ml with d istilled water. The medium was
filter sterilized (0.45 pm), aliquoted in small volumes and stored at
-20°C.
II.B.12.e. 8-Azaguanine (250 X Stock)
A stock of 8-Azaguanine (Sigma) was prepared to a final concentration of 5 mg/ml in d istilled water. The 8-Azaguanine was brought into solution by the dropwise addition of 0.1 N sodium hydroxide. The medium was filter sterilized (0.45 pm), and aliquots were stored at -20°C. 37
II. C. RESULTS
II. C.l. Enzyme-linked immunosorbent assay
Optimal conditions for ELISA were determined by performing checkerboard titrations. Disrupted EIAV was diluted in TEN buffer so as to contain 1000 ng, 750 ng, 500 ng, 300 ng, 250 ng, 200 ng, and
150 ng per 50 yl respectively. A 50 yl sample of each virus dilutions was added to each well in a row across the ELISA plate. Bound antigen was detected using a two fold dilution series of polyclonal
immune mouse serumthat formed the second dimension of the checkerboard titrations. An optimal antigen concentration of 1 ug/well of dissociated virus gave the highest results at 50% maximum optical readings as shown in Figure I I . 1. This concentration of antigen was used to maximize sensitivity to small quantities of
immunoglobulins that result from small colony size hybridomas and to increase the chance of detecting hybridomas specific to EIAV glycoproteins. Theoretically, ELISA plates were coated with about
100 ng/well of these glycoproteins since 10% of purified virus is composed of glycoproteins (120).
II.C.2. Isolation of hybrid cells producing anti-EIAV antibodies
Hybridomas which resulted from fusing SP2/0 myeloma cells and
B-lymphocyte cells were selectively isolated from the parental cells by growth in HAT medium in the wells of the microtiter plates.
Hybrid cells grew in each well that was seeded and culture fluids from approximately 4800 wells were examined for antibody activity against EIAV proteins using ELISA. A minimum of one hybrid was scored in each of the original wells (Fig. II.2). Cells from selected wells on the original plates which scored positively in ifrn cnetain o EIAV antigen. of concentrations different FIGU R E I I . 1. Titrations in ELISA of immune mouse serum against against serum mouse immune of ELISA in Titrations 1. . I FIGURE I
Absorbance at 490 nm 1.4 1.0 1.6 1.8 r .0 2 1.2 .2 4 ■ .4 .6 .8 0 - 0 ■
■
1.6 o 1 Dlto o Mue nieu t EIAV to Antiserum Mouse of Dilution 10 Log --1 2.2
2.8 3.4
0 ng 300 ------...... ------0 ng 200 .0 4 _i ______5 ng 750 5 ng 250 0 ng 500 00 ng 1000 "a: 5 ng 150 i_ 4.6 38 39
FIGURE I I .2. Early outgrowth of a hybridoma colony as a focus of relatively large, lymphoid cells in a background of thymocyte feeder cells, debris, and dying unfused lymphocytes ( 350X). 40
ELISA (15 to 20% of wells produced antiviral antibodies) were cloned
twice by limiting dilution and again retested for antibody activity
until they had achieved phenotypic stability. From the 12
independent fusions 150 hybridomas were obtained and ascitic fluid
was produced. The serological reactivity of these hybridomas was monitored by enzyme immunoassay (ELISA and Western blot immunoassay)
and neutralization test.
II.C .3. Proteins recognized by sera from immunized mice
Serum obtained from immunized mice was tested in Western blot
immunoassay with the proteins of EIAV resolved by SDS-PAGE (Fig.
I I .3). Serum obtained from mice immunized with glycoprotein
preparations had most of its activity against gp90, gp45, and pl5, while reacting weakly with p26. While serum from mice immunized with
dissociated virus plus adjuvant was mostly active against gp90, p26,
and pl5, it reacted faintly with gp45. In fact the reaction with
gp90 was as strong as the serum obtained from glycoprotein inoculated mice. In contrast, serum of mice inoculated with whole virus plus adjuvant reacted weakly with gp90 and gp45 but very strongly with p26 and pl5. The same results were obtained using serum from mice
immunized with whole virus given intravenously except that reactions
to pl5 were weaker in these mice. Surprisingly, serum from mice
immunized with whole virus (2 inoculations) with no intraperitoneal
inoculation reacted strongly with p26 only.
II.C.4. Identification of viral proteins detected by monoclonal antibodies
Monoclonal antibodies were examined by Western blot immunoassay with the prototype EIAV to check for their specificity. A total of abcdefgh i Jkl
gpW
■;W*'
gP«
p Z B
pis
FIGURE I I .3. Western blot analysis of serum samples from mice immunized with EIAV. Polypeptides of EIAV were separated on SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted with immune mouse serum. Lanes a, b and c are negative controls, (a) virus only; (b) conjugate only; and (c) normal mouse serum; (d) serum from mice immunized with glycoprotein; (e and f) serum from mice immunized with dissociated EIAV; (g and h) serum from mice immunized with whole EIAV; (i and j) serum from mice immunized intravenously with whole EIAV; and (k and 1) serum from mice immunized with whole EIAV with no intraperitoneal inoculation. 42
23 independent cloned hybrid cell lines that produced antibody against EIAV proteins were isolated and selected out of the 150 hybridomas in Western blot immunoassay (Table I I . 1). Ten of these
MCAbs were specific for gp90, 8 for gp45, and 5 for p26. MCAbs
86-1E3, 71-1A9, 95-1G8, 85-1E11, 87-1E7, 115-3D7, 98-101, 82-1C2,
114-3A7, and 128-2B9 were specific for gp90 in Western blot immunoassay (Fig. I I .4). MCAbs 75-1F2, 101-2F10, 92-1E6, 105-3C8,
90-1C1, 117-1C5, 109-1A6, and 120-1H9 were specific for gp45 (Fig.
11.5). The MCAbs 141-4D5, 140-1C3, 149-4E3, 135-4E5, and 131-2D6 reacted with the group specific antigen, p26 (Fig. II.6). These monoclonals showed reactivity in Western blot immunoassay with another protein of an apparent molecular weight of about 55,000 (Fig.
11.6).
II.C.5. Characterization of MCAbs
Immunoglobulins in the culture fluid from each of the hybrid cell lines were concentrated by ammonium sulfate precipitation (50% saturation) and tested in immunodiffusion assays with heterologous anti sera prepared against purified mouse immunoglobulins.
Immnoglobulins produced by hybrids 86-1E3, 95-1G8, 114-3A7, 71-1A9,
85-1E11, 82-1C2, 87-1E7, 98-1D1, 115-307, 75-1F2, 92-1E6, 101-2F10,
109-1A6, 117-1C5, 120-1H9, 90-1C1, 149-4E3, 141-4D5, 140-1C3, and
131-2D6 were of IgGl subclass (Fig. I I . 7). Immunoglobulins produced by 105-3C8 and 135-4E5 were of IgG2b subclass, and the immunoglobulin produced by the hybrid 128-2B9 was of an IgA sublass (Table I I . 1).
II.C.6. Protein A binding assay
Positive hybridomas were also tested for their ability to bind protein A. The results indicated that MCAbs 105-3C8 and 135-4E5, Table I I .1. Characterization of monoclonal antibodies to EIAV polypeptides
Clone No. Isotype Protein A binding Antigen specificity
86-1E3 IgGl + gp90 95-1G8 IgGl - gp90 114-3A7 IgGl + gp90 85-1E11 IgGl + gp90 71-1A9 IgGl + gp90 87-1E7 IqGl - gp90 82-1C2 IgGl + gp90 98-1D1 IgGl - gp90 115-3D7 IgGl + gp90 128-2B9 igA - gp90 75-1F2 IgGl + gp45 92-1E6 IgGl + gp45 101-2F10 IqGl + gp45 109-1A6 IgGl + gp45 117-1C5 IgGl + gp45 120-1H9 IgGl + gp45 105-3C8 IgG2b +++ gp45 90-1C1 IgGl + gp45 131-2D6 IgGl + p26, Pr55gag 135-4E5 IgG2b +++ p26, Pr55gag 140-1C3 IgGl + p26, Pr55gag 141-405 IgGl + p26, Pr55gag 149-4E3 IgGl + p26, Pr55gag
+++ denotes strong binding in ELISA (three standard deviations above the mean) and positive in protein blot immunoassay using protein A peroxidase. + low affinity in ELISA - negative 44
* 1 H
m
FIGURE I I .4. Western blot analysis of 10 anti-gp90 monoclonal antibodies. Immunoblots were performed as described in the text. The polypeptides of EIAV were separated by SDS-PAGE, and transferred to nitrocellulose membranes. The strips were incubated with monoclonal antibodies 86-1E3, 95-1G8, 114-3A7, 85-1E11, 71-1A9, 87-1E7, 82-1C2, 98-1D1, 115-3D7 and 128-2B9 in lane A through J, respectively. Lane K represents immune mouse serum, showing immunoreactivity with gp90 and gp45 of EIAV. 45
FIGURE I I . 5. Western blot analysis of 8 anti-gp45 monoclonal antibodies. Immunoblots were performed as described in the text. The strips were incubated with monoclonal antibodies 75-1F2, 109-1A6, 105-3C8, 92-1E6, 90-1C1, 101-2F10, 117-1C5, and 120-1H9 in lane A through H, respectively. Lane I represents immune mouse serum, showing immunoreactivity with gp90 and gp45 of EIAV. 4 6
a bed e f
94k *■" 67k
43k
■■; $M0M 14.4k
FIGURE I I . 6. Western blot analysis of 5 anti-p26 monoclonal antibodies. Immunoblots were performed as described in the text. The strips were incubated with monoclonal antibodies 131-2D6, 135-4E5, 140-1C3, 141-4D5, and 149-4E3 in lane b through f, respectively. Blot of protein standards stained with amido black (lane a) were included as reference. The arrow head shows the location of the 55K protein. 47
FIGURE I I . 7. Immunodiffusion analysis. Immunoglobulin product of selected hybridomas: 86-1E3, 82-1C2, 105-3C8, 98-1D1, 128-2B9, and 115-3D7 were arranged radially in individual peripheral wells 1 through 6, respectively while goat anti-mouse IgGl was plated in the center well. Monoclonal antibodies in wells 1, 2, 4 and 6 were positive (IgGl) while MCAbs in wells 3 and 5 were negative. 48
which are of IgG2b isotype, bind strongly to protein A. On the other
hand, MCAbs 92-1E6, 109-1A6, 90-1C1, 120-1H9, 75-1F2, 101-2F10,
86-1E3, 117-1C5, 149-4E3, 141-4D5, 140-1C3, 131-2D6, 114-3A7,
85-1E11, 115-3D7, 82-1C2, and 71-1A9 exhibited variable degrees of
binding to protein A. The MCAbs 95-1G8, 87-1E7, 98-1D1, and 128-2B9
were negative for protein A binding (Table I I . 1).
II.D. DISCUSSION
Hybridoma technology has created a revolution in the biomedical
sciences in general and virology in particular. It is possible using
MCAb procedures to produce large quantities of antibodies specific
for a single epitope on a single viral polypeptide using crude
antigen for immunization. These MCAbs are the most specific
biological probes available for the antigenic analysis of proteins.
Alterations of surface glycoproteins are believed to be involved in
antigenic changes observed in EIAV variants (129). The sensitivity
and specificity of MCAbs can be used to define and map specific
virion changes that contribute to the observed altered antigenicity
of EIAV variants in animals persistently infected with EIAV.
Therefore, the objectives of these studies were to generate a diverse
panel of hybridoma antibodies against mainly the surface
glycoproteins of EIAV to compare the glycoproteins of EIAV isolates
and to define the determinants involved in neutralization.
Elaborate protocols for immunization and cell hybridization exist in the literatu re, but there is no consensus on the most appropriate protocol. In devising an efficient fusion technique
several variables were taken into consideration. One of these
variables is the immunization schedule. The immunization scheme used 49
in this study has resulted in good immune responses in mice (ELISA
tite r 2 640). However, the most universally accepted procedure in
the variety of published protocols is that the final immunization,
either intravenous or intraperitoneal, be given 3 to 4 days before
the spleen is harvested for fusion. The other variable was the
selection of PEG, therefore, several batches of commercially
available PEG were tested, as some different batches have been
reported to be toxic to cells (48). The hybridization scheme is another important variable taken into account. Therefore, an appropriate hybridization scheme was selected and modified to achieve a good frequency of hybrid formation. It is clear that generating as diverse a panel of MCAbs as possible will help in dissecting the antigenic sites of the surface glycoproteins. Several factors must be considered if one aims at generating a diverse panel of MCAbs.
These factors were evaluated for anti-influenza virus hybridomas by
Yewdell and Gerhard, (154) and include (1) the number of individual mice used as antigen-primed lymphocyte donors (2) the immunization protocol, and (3) the organ source of the lymphocytes used for fusion. Some of these factors were considered in our studies for the production of MCAbs specific for glycoproteins of EIAV. To increase the chance of isolating hybridomas which produced antibodies to these glycoproteins, we used immunogens enriched in their gp90 and gp45 content following len til-lec tin column chromatography, since purified virus contains only about 10% of these glycoproteins by mass (120).
Different preparations of antigen were also inoculated into mice with the hope that a diverse panel of MCAbs would be generated. It is obvious from Western blot immunoassays that sera from mice immunized with different preparations showed different reactivities against
EIAV proteins. Serum from mice immunized with either dissociated
virus or glycoprotein preparations had the strongest reactivity against gp90 and gp45. In addition, serum from mice immunized with either whole or dissociated virus neutralized infectivity but with
varying potencies (neutralization tite r ranged from 80 to 320, Table
3, Chapter III). Previous studies (154) indicated that individual antisera obtained at different periods of immunization often contain distinct spectra of antibody specificities. To increase the diversity of EIAV specific hybridomas we performed several fusion experiments in this study and modified the time from initial
immunization to harvest splenic cells to maximize the diversity.
The screening of hybridomas for antibody production is one of the essential requirements of the fusion experiments. At present, the most difficult aspect of hybridoma technology is to identify and select hybridomas that are producing antibody of the desired specificity. Therefore, an appropriately designed sensitive screening scheme that permits early detection of specific antibodies is a necessity. However, the choice of screening assay for specific antibody depends on the nature of the antigen and its availability in pure form. A prime consideration in the choice of any assay is the ease, rapidity, and reproducibility for handling large numbers of samples. However, the best approach to screen hybridomas is to use methods that allow detection of the type of antibodies against antigens of interest. If, for example, one wishes to obtain antibodies that can be used to immunoprecipitate protein antigens from cell lysates, immunoprecipitation is probably the best screening assay to use. If one needs antibodies that bind antigen in Western
blot immunoassay, or that are cytotoxic in the presence of complement or function in cellular cytotoxicity assays, one should screen for antibodies with just those properties. The strategies followed for screening EIAV specific hybridomas was dictated by and correlated with the objectives of developing MCAbs, i.e., to generate as diverse an antibody panel as possible. Therefore, a reasonably sensitive
ELISA was developed to detect hybridomas against EIAV proteins.
However, ELISA was regarded as a screening assay, not as an absolute definition of antigen-antibody specificity. Therefore, ultimate determination of MCAbs reactivities to EIAV included Western blot immunoassay and neutralization test (Chapter III). The binding assay used here should not, therefore, be used to the exclusion of other more informative screening assays such as neutralization test.
Although MCAbs are directed at specific antigenic determinants, it is often of interest to understand the native conformation of the protein possessing these determinants. In that regard, determinants are often referred to as continuous, discontinuous, or topographic.
A continuous determinant is defined as a continuous sequence of residues exposed at the surface of a native protein and possessing distinctive conformational features. A discontinuous determinant consist of residues that are not contiguous in the primary structure but are juxtaposed through folding and formation of the secondary structure of the protein. However, the topographic determinant is described as an epitope which gains its structural uniqueness by long-range interactions at the level of secondary and tertiary structure. It is likely that MCAbs specific for continuous or 52 discontinuous epitopes on EIAV glycoproteins were detected by ELISA.
Soluble proteins have been shown to attach to solid phase by passive absorption to plastic without detectable influences on their antigenic structure (35). However, MCAbs that recognize topographically assembled epitopes may not be recognized in ELISA tests using deoxycholate disrupted virus and neutralization tests are often more relevant in this regard. Unfortunately such a neutralization assay which affords early screening for detection of neutralizing antibodies does not exist for EIAV at the moment.
The identification of viral proteins detected by MCAbs was determined by Western blot immunoassay. Western blot immunoassay combines the high resolving power of SDS-PAGE and the high sensitivity of the ELISA to produce an extremely powerful qualitative tool for determining the specificity of antibodies against the spectrum of EIAV polypeptides. The fact that the antigens in Western blot immunoassay areall treated with SDS, brings forth the theoretical consideration of loss of native configurational integrity of the potential antigenic epitopes. There is the danger that not all the antigenic sites can retain the native configuration after SDS treatment to allow recognition by the appropriate antibody. However, experiments with IgG, BSA, and other defined antigens, would tend to negate this argument, since transfer of polypeptides onto nitrocellulose membranes allows polypeptides to lose their SDS coating and may allow sufficient refolding of proteins into their native conformation to allow antibody recognition (142). Only further investigations can fully answer these questions. Western blot immunoassay proved to be rapid and efficient in screening of 53
these MCAbs. Numerous applications of MCAbs in basic research
require purification of MCAbs from the rest of the proteins in
ascitic fluids. The most conventional method to purify MCAbs is to
use protein A chromatography. Therefore protein A binding assays
will be helpful to identify MCAbs that bind protein A. The variation
exhibited by some EIAV specific MCAbs in protein A binding assay may
be simply attributed to variation in antigen concentration of
different EIAV polypeptides or the pH may not be optimal for the
their binding. It is known that the association constant for binding
of IgGl immunoglubulins to protein A is extremely pH dependent (37).
Serological reactivities of EIAV specific MCAbs as monitored by
Western blot immunoassay indicated that 10 MCAbs were specific for
gp90, 8 for gp45, and 5 for p26. Monoclonal antibodies which reacted
with p26, also reacted with another protein of an apparent MW of
55,000. The presence of multiple bands on the protein blot reflect
antigenic relatedness. This protein is most likely to be the gag
precursor of EIAV (referred to as Pr55gag). Previous studies
demonstrated the unrelatedness of the four major nonglycosylated
internal (gag) proteins of EIAV (p26, pl5, pll, and p9) by peptide mapping and serology (100). In addition the aggregate collective molecular weight of the gag proteins is equivalent to 61K, which
approximates that of their putative gag precursor (55K). These
observations along with the cross reactivity noted between p26 and
the 55K protein suggest that these proteins (p26, pl5, pll, and p9)
are cleavage products of the 55K protein (Pr55gag). However, the
epitope recognized by these MCAbs must be located on the group
specific antigen (p26) since these MCAbs showed reactivity with both the 55K precursor protein and its cleavage product, p26. On the other hand, all the glycoprotein specific MCAbs were highly specific
in their reactivity with either gp90 or gp45. Additionally, no protein of molecular weight higher than gp90 could be detected by our panel of MCAbs and no cross reactivity was observed between gp90 and gp45. Both of these observations argue against the existence of a higher molecular weight protein that might be a common precursor of the envelope glycoproteins. These observations demonstrated the unrelatedness of these surface glycoproteins and proved that they are distinct.
The battery of MCAbs generated for the surface glycoproteins
(gp90 or gp45) of EIAV represent valuable reagents which may have tremendious benefit in the identification of epitopes necessary for protective immunogenicity. CHAPTER III
ANTIGENIC ANALYSIS OF EQUINE INFECTIOUS ANEMIA VIRUS VARIANTS BY
USING MONOCLONAL ANTIBODIES: EPITOPES OF GLYCOPROTEIN 90 (GP90) OF
EIAV STIMULATE NEUTRALIZING ANTIBODIES
III.A. INTRODUCTION
Equine infectious anemia virus, a lentivirus, causes an
important disease in all members of the horse family. The disease
EIA is characterized by viral persistence, immunologically mediated
lesions, and a variable clinical course (29, 60). Biochemically, the
virus contains two main surface glycoproteins (gp90 and gp45) and four major nonglycosylated internal proteins (100, 120). One possible mechanism for viral persistence with EIAV involves the evolution of antigenic variants which allow the virus to temporarily elude established host immune defenses. Antigenic variation in EIAV
is thought to involve frequent point mutations of the viral genetic material in those specific areas coding for the viral surface proteins. In an infected animal, such an alteration of the virus antigenic structure would allow the emerging variant to escape from neutralization by preexisting antibody, thus, perpetuating the virus
infection in an immunologically competent host. The evolution of these antigenic variants which are not susceptible to previously formed antibody may explain the periodic febrile relapses and the persistence of viremia in horses infected with EIAV.
Antigens displayed on the surface of the virion serve as potential targets for neutralizing antibodies (99, 130). Variants of the lentiviruses EIAV and visna virus contain alterations confined to the envelope glycoproteins (22, 102, 129, 132). In this instance,
55 56 the emerging mutant viruses have altered envelope glycoproteins, and these alterations are purportedly sufficient to allow the virus to temporarily escape immunological inactivation. Although i t would be premature to predict the role of antigenic variation in human immunodeficiency virus (HIV), isolates of this virus from different individuals differ genotypically, and this genomic heterogeneity is greatest in the region of the env gene (6, 125, 152). Recently, Hahn et a l. (49) suggested that human immunodeficiency viruses isolated sequentially from persistently infected individuals have evolved in parallel from a common progenitor virus. This type of variation may be common to all members of the subfamily Lentivirinae. If these genotypic and phenotypic changes are mirrored antigenically, they will undoubtedly pose serious problems for vaccine development and provide an essential mechanism for these viruses to persist and spread in individuals and populations.
Monoclonal antibodies have proven to be invaluable in studying antigenic variation of influenza viruses as well as many other viruses. The number and location of antibody binding sites on the hemagglutinin (HA) and neuraminidase (NA) were determined using MCAbs in selecting variants and in competitive binding assays which demonstrated the presence of three or four nonoverlapping antibody binding sites (10, 144, 148).
In addition, the use of MCAbs for analyzing the epitopes on viral proteins represents a very important new tool for identifying critical site(s) involved in virus neutralization. The discriminatory power of MCAbs make them ideal reagents for investigation of antigenic relatedness between viral proteins. The alterations of virus antigenicity, detectable only by MCAbs, seem to
be of major relevance in virus host interactions. Coxsackievirus
Type B4 is an example where antigenic variants of the virus were found to be different from the parental strain, and from one another,
in their myocarditic and cardiotropic properties in a murine model
(14). Therefore, we were committed to the production of MCAbs to study antigenic variation of EIAV, and to define epitopes alteration
in virion surface glycoproteins. Through the use of a diverse panel of MCAbs it will be possible to dissect or map out the altered antigenicity of the surface glycoproteins that signal the emergence of antigenically distinct variants capable of eluding the immune defenses. These data will provide additional information on alteration of antigenicity that may parallel the structural and genomic variation observed with EIAV variants (102, 121). In addition, MCAbs against the EIAV surface glycoproteins will help to
identify antigenic determinants important for neutralization. The surface glycoproteins of selected EIAV variants will be compared using a battery of neutralizing MCAbs to map out the protective antigenic determinant(s). It may help also to identify determinants common to all EIAV variants that may afford cross protection.
III. B. MATERIALS AND METHODS
III. B. 1. Virus production, propagation, and purification
All virus isolates used for this study were produced by serial passage of prototype strain of EIAV in Shetland ponies (116). Virus isolates were designated by the letter "P" followed by a three-digit number with the first, second, and third digit corresponding to the passage number, pony number, and isolate number, respectively (e.g., PI.1-1). The prototype cell-adapted Wyoming strain of EIAV (88) and
isolates P2.1-1, P2.1-6, P3.1-1, P3.1-2, P3.1-3, P3.1-4, P3.2-1,
P3.2-2, P3.2-3, P3.2-4, P3.2-5, P3.3-1, P3.3-2S P3.3-3, and P3.3-4
(116, 117, Rwambo et a l ., submitted) were propagated in primary FEK cells and purified according to described procedures in chapter II.
The virus isolates obtained from plasma collected during the
indicated sequential febrile episodes of experimentally infected
Shetland ponies were purified by end point dilution in FEK cells to ensure isolation of the predominant virus population (102). Isolates
P2.1-1 and P2.1-6 were obtained from an experimentally infected pony during the first and sixth febrile episodes (Fig. 111.1). The remaining 13 isolates were recovered from 3 Shetland ponies infected with the same virus inoculum (1 ml of plasma collected from pony #82 during its first febrile episode, P2.1-1, with plasma infectivity of 4 5 10 * 50% tissue culture-infective dose [TCIDjjq] per 0.5 ml). These
isolates are identified as follows: 4 consecutive isolates (P3.1-1 through P3.1-4) from the third passage animal #127 with each febrile episode separated by only 4 to 8 weeks; 5 consecutive isolates
(P3.2-1 through P3.2-5) from the third passage animal #91; and 4 consecutive isolates (P3.3-1 through P3.3-4) from the third passage animal #F135, representing consecutive febrile episodes separated by only 2 weeks to 4 months.
III. B. 2. Enzyme-linked immunosorbent assay (ELISA)
The procedure used as a screening assay was as described in
Chapter II. Ascitic fluids were produced for positive selected clones and were serially diluted in TEN buffer to determine their reactivity in ELISA. 59
PROTOTYPE STRAIN
EIAV
First passage: P I.1-1 (pony #47) I Second passage: P2.1-1 ...... P2.1-6 (pony #82) \ P3.1-1, P3.1-2, P3.1-3, P3.1-4 (pony #127)
Third passage: / y
P3.2-1, P3.2-2, P3.2-3, P3.2-4, P3.2-5 (pony #91)
P3.3-1, P3.3-2, P3.3-3, P3.3-4 (pony #F135)
FIGURE 111.1. Inoculation scheme followed for the production of EIAV isolates in experimentally infected Shetland ponies. Each of the three ponies (#127, #91, and #F135) received 1 ml of plasma collected from pony #82 during its first febrile episode. Plasma infectivity was 10 * TCIDjjq/ 0.5 ml. 60
III. B. 3. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis
(SDS-PAGE)
Sodium dodecyl sulfate-polyarylamide gel electrophoresis was
performed in slab gels of 18 cm X 32 cm X 1.5 mm according to Laemmli
procedure (79). Gels used were 10% polyacrylamide containing 0.2%
bis-acrylamide, 0.1% SDS, 0.1% TEMED (N, N, N 't N'-
Tetramethylethylenediamine), 1 M urea, and 100 mM sodium phosphate buffer, pH 7.2. Gels were polymerized by the addition of 4 ml of 1% ammonium persulfate per 76 ml of acrylamide monomer. Electrophoresis buffer consisted of 100 mM sodium phosphate, pH 7.2, with 0.1% SDS and 10 mM 3-mercaptopropionic acid. All gels underwent electrophoresis for at least 1 h at normal running conditions prior to application of samples in order to eliminate any peroxides formed during the polymerization reaction. The procedures used for sample preparation and SDS-PAGE in continous sodium phosphate buffer for
EIAV have been previously described (100). Briefly, aqueous samples of unlabeled EIAV (60 yg) were brought to a volume of 90 yl with 10 mM sodium phosphate buffer, pH 7.2. Sodium dodecyl sulfate and
2-mercaptoethanol were added to achieve a 1% concentration for each in the sample, and heated for 3 min in a boiling water bath. Sucrose was added to give a final concentration of 20 to 30%, and 1 yl of
0.5% bromophenol blue in 0.02 N sodium hydroxide was added to each sample as a tracking dye. Samples were layered onto the gel followed by electrophoresis at 50 mA per slab gel for 20 to 24 hr. Buffer was circulated between the upper and lower chambers to prevent large pH differences from developing during electrophoresis. After separation by SDS-PAGE, proteins were either visualized in slab gel by staining 61 in 0.25% coomassie b rillia n t blue in acetic acid:methanolrwater
(46:227:227, V/V/V) for 1 h with continuous agitation, followed by destaining in acetic acid:methanol:water (3:10:27, V/V/V), or transferred to nitrocellulose membrane.
III.B.4. Western blot immunoassay ( Immunoblotting)
The electrophoretic transfer of proteins after separation by
SDS-PAGE to nitrocellulose membranes was adapted from the method of
Burnette (13) and Towbin and Gordon (143). Briefly, f ilte r papers, sponge pads and nitrocellulose membranes (Bio-Rad Laboratories,
Richmond, California) were soaked for 10 min in a transfer buffer consisting of 25 mM Tris, 192 mM glycine, and 20% methanol, pH 8.3.
The SDS gel containing separated proteins was washed immediately after electrophoresis with transfer buffer, then equilibrated in the same buffer for 30 min. The gel was placed in direct contact with the nitrocellulose membrane and sandwiched between two pieces of
Whatman 3 mm filter papers and the polyethylene sheets of the cassette. The cassette was placed in an electroblot unit (Bio-Rad
Laboratories), with the nitrocellulose oriented toward the anode.
The unit was filled with cold transfer buffer. Electrophoretic transfer was performed at 15°C for 18 to 20 h at a constant current of 0.25 A. The nitrocellulose membrane with transferred proteins was dried for 1 h at 37°C before being used for immunoblotting.
Immunoreactive components were identified by Western blot immunoassay according to the procedures of Towbin and Gordon (143) and Hawkes
(50). The nitrocellulose membranes were immersed in 3% bovine serum albumin (Sigma Chemical Co., St. Louis, M0) in 10 mM Tris hydrochloride (pH 7.4) with 150 mM sodium chloride for 30 min at room temperature with continuous agitation. The membrane was incubated for 90 min with gentle shaking at room temperature in a 1:100 dilution of ascitic fluid or a reference antiserum from a naturally infected horse. The nitrocellulose membranes were washed for 10 min in Tris-saline followed by two 10 min washes in Tris-saline containing 0.05% NP-40, with gentle agitation. The nitrocellulose membrane was then incubated for 1 h in Tris-saline containing a
1:1000 dilution of peroxidase conjugated goat anti-mouse immunoglobulins. The nitrocellulose membrane was washed as described above, and color developed by incubating for a few minutes at room temperature with the color development solution (4-chloro-l-naphthol; chapter II). The development was stopped by immmersing the membrane in distilled water for 10 min.
III.B.5. Neutralization of viral infectivity with monoclonal antibodies (Beta Procedure)
Monoclonal antibodies and immune mice sera were evaluated for their ability to neutralize the cytopathic activity of EIAV. Samples to be used in neutralization assays were filtered (0.45 yin) and heated at 56°C for 30 min before mixing with the virus. Serial two-fold dilutions of ascitic fluids were mixed with an equal volume
(50 yl) of virus suspension (100 TCID^q ) in MEM and incubated for 1 h at 37°C. Cells derived from fetal donkey dermal cultures
(10,000/well) were added in 100 yl volume to each well as indicators for EIAV infectivity. Incubation was carried out for 7 to 12 days, and cells were observed for cytopathic effect (CPE). The neutralization titer was expressed as the reciprocal of the highest dilution which inhibited the cytopathic effect of 100 TCIDcr, of the ou 63
EIAV in fetal donkey dermal cell cultures.
III.B.6. Neutralization activity (Alpha Procedure)
The neutralization activity of anti-gp90 monoclonal antibodies was evaluated in a neutralization index assay using a constant serum-varying virus assay. Sera and ascitic fluids to be tested for antibody content were heat inactivated at 56°C for 30 min before mixing with the virus dilution. A constant volume of MCAb or serum was reacted with equal volumes of varying virus dilutions (10 * to
10”^) of EIAV for 1 h at 37°C. Next, 0.5 ml of each mixture was adsorbed to FEK monolayer cell cultures (grown in 25 cm tissue culture flasks) for 1 h at 37°C. After the incubation period, each flask was rinsed with PBS and refed 5 ml of maintenance medium (MEM with 3% FCS) and incubated further at 37°C for 2 to 3 weeks. Virus titrations were done simultaneously andall assays were done in duplicate. When the neutralization assay was performed with complement, the same procedure was used as outlined above except that the mixture was incubated for 30 min at 37°C instead of 1 h. After th is, 50 yl of guinea pig complement (pooled guinea pig serum from
Miles Laboratories, Inc.) was added and the virus-antiserum or ascitic fluid complement mixtures were incubated for an additional 30 min at 37°C. The inoculum was then adsorbed to FEK monolayer for 1 h at 37°C and the above described procedure was again followed. Virus neutralization was evaluated by detecting the presence of p26 antigen
in the supernatant fluid in agar gel immunodiffusion assay since EIAV does not produce CPE in these cultures (1). The neutralizing activity of the serum or ascitic fluid was expressed as a log^g neutralizing index, calculated as the difference in log^Q of the tite r of the virus samples with or without te st serum. A neutralizing index of £ 1.7 was regarded as positive, i .e ., neutralization.
III.C. RESULTS
III.C.l. Western blot analysis
Western blot analysis of 7 EIAV isolates using our reference immune serum was performed simply to identify their polypeptides as four of these sequential virus isolates (P3.1-1, P3.1-2, P3.1-3, and
P3.1-4) have proven to be antigenically distinct (Table 111.1) by cross-neutralization studies (117). The polypeptides of these isolates and the prototype strain, were resolved by SDS-PAGE in 10% gels (Fig. III.2). Although the two major surface glycoproteins cannot be easily located in this electropherogram, the nonglycosylated internal proteins (p26, pl5, pll, and p9) can be identified in all seven isolates. The resolved proteins of EIAV on
SDS-PAGE were electrophoretically transferred to nitrocellulose membrane and immunoblotted with reference horse immune serum.
Protein standards as well as prototype EIAV proteins bound to nitrocellulose membranes were stained with ami do black to monitor their electrophoretic migration and transfer. The immunoblot (Fig.
111.3) demonstrated that both gp90 and gp45 of the prototype EIAV were intensely recognized by the reference anti serum, however, less reactivity against p26 was observed even though p26 is present in much higher quantities than the glycoproteins (about 5 fold). When the reference immune serum was reacted with the glycoproteins of seven isolates variation in theimmunoreactivity was observed. The serum reacted weakly with gp45 of isolates P2.1-6, P3.1-2, and 65
TABLE III.1. Serum neutralization indices of three sera collected from pony #127 at different times after inoculation. Each serum was reacted with sequential virus isolates from the same pony.
2 Neutralization index of serum collected
on the following day postinoculation
Virus isolate Day PI1 31 125 158
P3.1-1 13 1.5 3.0 3.0
P3.1-2 41 0.5 3.0 2.5
P3.1-3 99 1.5 2.0 2.5
P3.1-4 137 1.0 1.0 1.0
Days postinoculation when plasma was collected. 2 Log^Q serum-neutralization index is the difference in Log^g of the
tite r of the virus samples with or without test serum (117). 66
\ h few* 94k p d |*Ǥ **""** |0 6 7 k
*»(■*<•§i*“t ..P.«i,«. '*!'
i i ■.= ■■■
k*$)SS “!V“ W 43k -Mi ; ¥ -f
y i3 0 k
p 2 6 %
20-1k
p15( p11 P 9
FIGURE III,,2. Analysis of EIAV polypeptides. SDS-PAGE separation of virion proteins (visualized by coomassie blue staining) of gradient-purified EIAV isolates: prototype (a), p2.1-l(b), p2.1-6(c), p3.1-l(d), p3.1-2(e), p3.1-3(f), p3.1-4(g), and protein standards (h). 67
a b cd e f 9
HMl94k
|^|g<7k
43k gp45 if
% | 3 0 k p 26 *
201k
FIGURE I I I .3. Western blot analysis of reference horse immune serum with seven EIAV isolates. Polypeptides of prototype (a), P2.1-1 (b), P2.1-6 (c), P3.1-1 (d), P3.1-2 (e), P3.1-3 (f), P3.1-4 (g) were separated on SDS-PAGE (similar to Fig. 1), transferred to nitrocellulose membrane, and immunoblotted with reference antiserum. Equal amounts (60 yg) of each virus isolate are present in each lane. Prototype EIAV (h) and protein standards (i) were included as references and stained with ami do black. 68
P3.1-4, whereas it reacted strongly with gp45 of all other isolates.
In addition, variation in electrophoretic mobility of both gp90 and gp45 from all seven isolates was evident suggesting structural alterations might have occurred in these two glycoproteins. To identify the contributions of both gp90 and gp45 to the antigenic variation, these isolates as well as other sets of isolates obtained from a parallel infection were subjected to antigenic analysis using a panel of 18 monoclonal antibodies.
III.C.2. Reactivity of monoclonal antibodies with the homologous virus strain
The cell-adapted Wyoming strain of EIAV was used to inoculate mice for the production of MCAbs as described in detail in Chapter
II. Monoclonal antibodies produced using different immunizing schemes were compared (Table I I I .2). Of the 150 hybridomas producing
MCAbs produced to EIAV, 18 reacted with the glycoproteins of the prototype EIAV by Western blot immunoassay. Four of the 18 MCAbs had neutralizing activity with titers ranging from 64 to 128. Indirect
ELISA tite r of ascitic fluids ranged from 1 X 10^'^ to 1 X 10^*®.
The most frequent isotype was IgGl followed by IgG2b and IgA (Table
I I I .2).
III.C.3. Reactions of monoclonal antibodies with a panel of EIAV variants
A total of 18 independent stable hybridomas reacting to EIAV glycoproteins (gp90 or gp45) were selected for this study. Each of these MCAbs was tested in Western blot immunoassay against a panel of
16 EIAV variants. All virus variants obtained during the indicated febrile episodes were purified by end point dilution in FEK cells to Table III.2. Characterization of anti-EIAV Monoclonal antibodies
MONOCLONAL ANTIBODY REACTIVITY TO EIAV
Clone No. Isotype Immunizing antigen Viral protein ELISA Neutralization immunoblotted titer (log1Q) (titer)
86-1E3 IgGl Disrupted virus gp90 3.9 95-168 IgGl Disrupted virus gp90 4.2 - 114-3A7 IgGl Disrupted virus gp90 3.9 - 85-1E11 IgGl Disrupted virus gp90 3.6 - 71-1A9 IgGl Glycoproteins gp90 3.3 - 87-1E7 IgGl Disrupted virus gp90 3.3 - 82-1C2 IgGl Disrupted virus gp90 3.6 +(128) 98-1D1 IgGl Disrupted virus gp90 3.3 + 64) 115-3D7 IgGl Disrupted virus gp90 3.6 +(128) 128-2B9 igA Whole virus gp90 3.6 +(128) 75-1F2 IgGl Glycoproteins gp45 3.6 - 92-1E6 IgGl Disrupted virus gp45 4.5 - 101-2F10 IgGl Disrupted virus gp45 4.5 - 109-1A6 IgGl Disrupted virus gp45 4.8 - 117-1C5 IgGl Whole virus gp45 3.9 - 120-1H9 IgGl Whole virus gp45 4.8 - 105-3C8 IgG2b Disrupted virus gp45 4.5 - 90-1C1 IgGl Disrupted virus gp45 4.5 - 131-2D6 IgGl Whole virus p26 3.6 - 135-4E5 IgG2b Whole virus p26 3.0 - 140-1C3 IgGl Whole virus p26 4.8 - 141-4D5 IgGl Whole virus p26 3.6 - 149-4E3 IgGl Whole virus p26 4.2 _
^■Reciprocal of highest dilution of ascitic fluid that inhibited the cytopathetic effect of 100 TCID™ of EIAV inoculum in fetal donkey dermal cell cultures. < 1:4 dilutions of ascitic fluid. 70
insure isolation of the predominant virus population (Fig 1II.1).
Eight patterns of immunoreactivity were observed, six for gp90 and
two for gp45. Each of these immunoreactivity patterns was assigned
to react with a putative epitope (epitopes 90-A, 90-B, 90-C, 90-D,
90-E, and 90-F for gp90-specific MCAb). Similarly epitopes 45-A and
45-B were assigned for gp45-specific MCAbs. These results can be
summarized as follows.
Antibodies from hybrids 86-1E3, 95-1G8, and 114-3A7 had a common
pattern of reaction and reacted with all isolates tested (Fig III.4).
Antibodies from hybrid 87-1E7 had a unique pattern of reaction.
This antibody reacted with the prototype EIAV, variants P2.1-1,
P3.1-1, P3.1-2, P3.1-3, P3.2-2, P3.2-3, P3.3-2 and P3.3-3 and
completely failed to react with variants P2.1-6, P3.1-4, P3.2-1,
P3.2-4, P3.2-5, P3.3-1, and P3.3-4 (Fig. III.5).
Antibodies from hybrids 82-1C2 and 98-1D1 (neutralizing monoclonals) reacted with the prototype EIAV, P2.1-1, P3.1-1, P3.1-2,
P3.2-2, P3.2-3, P3.3-1, P3.3-2, and P3.3-3 but completely failed to
recognize P2.1-6, P3.1-3, P3.1-4, P3.2-1, P3.2-4, P3.2-5, and P3.3-4
(Fig. III.6).
Antibodies from hybrid 115-3D7 (neutralizing monoclonal) reacted with the prototype EIAV, P2.1-1, P3.1-1, P3.1-2, P3.2-1, P3.2-2,
P3.2-3, P3.3-1, P3.3-2, and P3.3-3 and failed to react with P2.1-6,
P3.1-3, P3.1-4, P3.2-4, P3.2-5 and P3.3-4 (Fig. III.7).
Antibodies from hybrid 128-2B9 (neutralizing monoclonal) reacted with the prototype EIAV, P2-1, P3.1-2, P3.1-4, P3.2-4 and P3.2-5 and failed to recognize P2.1-6, P3.1-1, P3.1-3, P3.2-1, P3.2-2, P3.2-3,
P3.3-1, P3.3-2, P3.3-3 and P3.3-4 (Fig. III.8). 71
FIGURE I I I .4. Western blot analysis of gp90 specific monoclonal antibodies with polypeptides of sixteen EIAV isolates. Polypeptides of prototype (a), P2.l-l(b), P2.1-6(c), P3.1-1(d), P3.1-2(e), P3.1-3(f), P3.1-4(g), P3.2-1(h), P3.2-2(i), P3.2-3(j), P3.2-4(k), P3.2-5(1), P3.3-l(m), P3.3-2(n), P3.3-3(o), and P3.3-4(p) viruses were separated on SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with monoclonal antibody 86-1E3. bcdefghijklmnop
FIGURE I I I .5. Western blot analysis of gp90 specific monoclonal antibody with polypeptides of sixteen EIAV isolates. Polypeptides were separated on SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with monoclonal antibody 87-1E7. EIAV isolates are in the same order as in FIGURE I I I .4. 73
FIGURE I I I .6. Western blot analysis of gp90 specific monoclonal antibodies with polypeptides of sixteen EIAV isolates. Polypeptides were separated on SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with monoclonal antibody 82-1C2. EIAV isolates are in the same order as in FIGURE I I I .4. 74
FIGURE I I I .7. Western blot analysis of gp90 specific monoclonal antibody with polypeptides of sixteen EIAV isolates. Polypeptides were separated on SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with monoclonal antibody 115-3D7. EIAV isolates are in the same order as in FIGURE I I I .4. 75
a fe c d t f 9 h I j k I m n o p
gp0O
FIGURE I I I .8. Western blot analysis of gp90 specific monoclonal antibody with polypeptides of sixteen EIAV isolates. Polypeptides were separated on SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with monoclonal antibody 128-2B9. EIAV isolates are in the same order as in FIGURE I I I .4. 76
Antibodies from hybrids 71-1A9 and 85-1E11 reacted similarly.
These antibodies reacted with the prototype EIAV, variants 3.1-3,
P3.2-3, P3.3-2 and P3.3-3 but failed to react with P2.1-1, P2.1-6,
P3.1-1, P3.1-2, P3.1-4, P3.2-1, P3.2-2, P3.2-4, P3.2-5, P3.3-1 and
P3.3-4 (Fig. III.9).
Antibodies from hybrids 75-1F2, 92-1E6, 101-2F10, 109-1A6,
117-1C5, 120-1H9, and 105-3C8 reacted with gp45 of all isolates tested (Fig. I I I . 10).
Antibodies from hybrid 90-1C1 reacted with gp45 of the prototype
EIAV, variants P2.1-1, P2.1-6, P3.2-1, P3.2-2, P3.2-3, P3.2-4,
P3.2-5, P3.3-2, P3.3-3, P3.3-4 and failed to react with variants
P3.1-1, P3.1-2, P3.1-3, P3.1-4, and P3.3-1 (Fig. I I I . 11).
III.C.4. Neutralization of EIAV infectivity by monoclonal antibodies
Since we were interested in developing neutralizing MCAbs against EIAV we injected mice with different preparations of the virus. Immune serum of EIAV recipient mice was evaluated for its neutralization capability (Table III.3). The results showed that both native EIAV and dissociated virus solubilized with deoxycholate induced high tite r antibody by neutralization assay. Therefore, the spleens of these mice were chosen as the doner for B lymphocytes for fusion with SP2/0 myeloma cells. Since the glycoproteins (gp90 and gp45) are located in the envelope of EIAV and represent the two main surface proteins, it was of interest to examine the MCAbs for neutralizing activity. Although all 18 of the MCABs bound to either gp90 or gp45 in Western blots, only 4 of these neutralized the infectivity of the virus to an appreciable titer (Table III.3 and
III.4). Each was directed against an epitope of gp90. The four FIGURE I I I . 9. Western blot analysis of gp90 specific monoclonal antibodies with polypeptides of sixteen EIAV isolates. Polypeptides were separated on SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with monoclonal antibody 85-1E11. EIAV isolates were in the same order as in FIGURE I I I .4. 78
a bed ef gh ! J k I
FIGURE I I I . 10. Western blot analysis of gp45 specific monoclonal antibodies with polypeptides of sixteen EIAV isolates. Polypeptides were separated on SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with monoclonal antibody 105-3C8. EIAV isolates are in the same order as in FIGURE I I I .4. FIGURE I I I . 11. Western blot analysis of gp45 specific monoclonal antibody with polypeptides of sixteen EIAV isolates. Polypeptides were separated on SDS-PAGE, transferred to nitrocellulose membrane, and immunoblotted with monoclonal antibody 90-1C1. EIAV isolates are in the same order as in FIGURE I I I .4. 80
TABLE I I I .3. Neutralization of EIAV by monclonal and polyclonal antibodies.
1 2 Sample tested Antigen recognized Neutralization Titer3
82-1C2 gp90 128
98-1D1 gp90 64
115-3D7 gp90 128
128-2B9 gp90 128
Mouse anti serum (Dissociated virus) gp90,gp45,p26,pl5 320
Mouse antiserum (Whole virus) gp90,gp45,p26,pl5 240
Mouse antiserum (Whole virus I/V) gp90,gp45,p26,pl5 80
■^Immunization schemes used to obtain immune mice sera were described in the text. 2 Antigen recognition was determined by immunoblot (see text for d e ta ils). 3 Reciprocal of highest dilution that inhibited the cytopathic effect of 100 TCIDrn of the EIAV inoculum in fetal donkey dermal cell cultures. 81
TABLE III.4. Neutralization indices of reference positive horse serum
and monoclonal antibodies.
Neutralization index of sample*
Sample Without Complement With Complement
Immune horse serum 2.5 2.5
82-1C2 2.0 2.0
98-1D1 2.5 2.5
115-3D7 2.5 2.5
128-2B9 2.0 2.0
^Log^Q Neutralization index is the difference in log^Q of the titer
of the virus with or without test sample. 82
MCAbs that had neutralizing activity (82-1C2, 98-1D1, 115-3D7, and
128-2B9) reacted with gp90 in three different patterns. These data demonstrated that MCAbs were directed against different neutralization epitopes on gp90 (epitopes 90-C, 90-D, and 90-E). The neutralizing ability of ascitic fluids and sera were also evaluated in the presence of a complement source to see whether complement enhanced the neutralization of sensitized virus. The results show that the addition of complement had no effect on the neutralization induced by all four MCAbs or by immune horse serum.
III.C.5. Comparison of 16 EIAV isolates
Monoclonal antibodies reacting to EIAV gp90 or gp45 were tested in Western blot assays against a panel of 16 EIAV variants (Fig.
III.1) to identify the pattern of glycoprotein antigenic variation.
Variation in electrophoretic mobility of both gp90 and gp45 from several isolates was evident when immunobloted with a given MCAb
(Fig. I I I . 12), as reported previously (102, 129). A serological comparison of the viral glycoproteins of all 16 EIAV isolates is summarized in Figure 111.13. Six patterns of immunoreactivity were observed for gp90 and two for gp45 when all 18 MCAbs were tested in
Western blot assays. Each of these immunoreactivity patterns could be assigned to a potential epitope (epitopes 90-A, 90-B, 90-C, 90-D,
90-E, and 90-F for gp90-specific MCAbs). Similarly epitopes 45-A and
45-B were assigned for gp45-specific MCAbs. Conserved epitopes were identified among all 16 virus isolates (epitopes 90-A and 45-A).
Each of these conserved epitopes was found independently on gp90 and gp45. The MCAbs that reacted with all EIAV isolates failed to neutralize the infectivity of the prototype strain. All MCAbs p 26
p15
FIGURE III.12. Western blot analysis of EIAV: polypeptides of prototype (lanes a, b, and d) and isolate P2.1-1 (lanes c and e) viruses were separated by SDS-PAGE, transferred to nitrocellulose membrabe, and immunoblotted with serum from mice immunized with disrupted virus preparations (a), neutralizing monoclonal antibody 82-1C2 (b and c), or monoclonal antibody 105-3C8 (d and e). m
VIRUSES EPITOPE P2.1 P3.1 P3.2 P 3.3 rn 1 6 1 2 3 4 1 2 3 4 512 3 4 9 0 -A ■■■ ■■■■■■■■■■■■■ 90-B 9 0-C 90-D 90-E □■□■□□□naiiinn 9 0 -F
45-A ■■■■■■■■■■■■■■■■ 45-B
FIGURE III.13. Immunoreactivity pattern of gp90- and gp45-specific monoclonal antibodies with polypeptides of 16 EIAV isolates. Polypeptides of the prototype (P), P2.1-1, P2.1-6, P3.1-1, P3.1-2, P3.1-3, P3.1-4, P3.2-1, P3.2-2, P3.2-3, P3.2-4, P3.2-5, P3.3-1, P3.3-2, P3.3-3, and P3.3-4 viruses were separated by SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted with each of the following MCAbs: 86-1E3, 95-1G8, and 114-3A7 (90-A); 87-1E7 (90-B); 82-1C2 and 98-1D1 (90-C); 115-3D7 (90-D); 128-289 (90-E); 85-1E11 and 71-1A9 (90-F); 75-1F2, 92-1E6, 101-2F10, 109-1A6, 117-1C5, 120-1H9, and 105-3C8 (45-A); and 90-1C1 (45-B). 85 specific for gp90 and gp45 reacted strongly with the respective glycoproteins from the prototype EIAV, but none of the isolates reacted with all the MCAbs. The immunoreactivity pattern exhibited by prototype EIAV was unique. Most isolates could be distinguished antigenically on the basis of their reactivity with the panel of glycoprotein specific MCAbs. Only isolates P3.2-4 and P3.2-5 and isolates P3.3-2 and P3.3-3 could not be differentiated.
A classification scheme for these EIAV isolates based on their immunoreactivity pattern was constructed (Table III.5). Nine different groups were noted on the basis of reaction with gp90 specific MCAbs. The most common are those of Groups 7 and 2, which are represented by three isolates each. Isolates P3.2-4 and P3.2-5 of Group 7 were recovered from the same animal and were identical for both gp90 and gp45 reactivity with this panel of MCAbs. They could, however, be distinguished from the other member of Group 7 (P3.1-4) because of specific changes in reactivity of gp45. In addition viruses belong to Groups 3 and 4 which have similar immunoreactivity with gp90 specific MCAbs, differ in their gp45 reactivity. In contrast Groups 2 and 9 have similar gp90 and gp45 immunoreactivity pattern. Only two immunoreactivity patterns were seen for the gp45 specific MCAbs. The first group contains the majority of isolates
(11 out of 16). Seven out of eight gp45 specific MCAbs recognized a highly conserved epitope on all 16 EIAV isolates (epitope 45-A). A second epitope on gp45 (epitope 45-B) appeared to be unique for the prototype and some other variants, e.g., four sequential isolates from pony #127 (P3.1-1, P3.1-2, P3.1-3, and P3.1-4). Four of the five isolates in Group 2A (P3.1-1, P3.1-2, P3.1-3 and P3.1-4) were 86
Table III.5. Classification of EIAV isolates by gp90 or gp45 immunoreactivity.
A. Classification by gp90 immunoreactivity pattern
GROUP 1 GROUP 2 GROUP 3 GROUP 4 GROUP 5 GROUP 6 GROUP 7 GROUP 8 GROUP 9 prototype P3.2-3 P2.1-1 P3.1-1 P3.1-3 P3.3-1 P3.1-4 P3.2-1 P2.1-6 P3.3-2 P3.1-2 P3.2-2 P3.2-4 P3.3-4 P3.3-3 P3.2-5
B. Classification by gp45 immunoreactivity pattern
GROUP 1A GROUP 2A
prototype P3.1-1 P2.1-1 P3.1-2 P2.1-6 P3.1-3 P3.2-1 P3.1-4 P3.2-2 P3.3-1 P3.2-3 P3.2-4 P3.2-5 P3.3-2 P3.3-3 P3.3-4 recovered from the same animal. Their gp45 had common changes, but they can all be differentiated on the basis of their gp90 reactivity.
Monoclonal antibody from hybridoma 90-1C1 reacted with prototype,
P2.1-1, P2.1-6, P3.2-1 through P3.2-5 and P3.3-2 through P3.3-4, but not isolates P3.1-1, P3.1-2, P3.1-3, P3.1-4 and P3.3-1. Only the first isolate P3.3-1 from the third animal had a loss of this epitope. This epitope reappeared in isolates P3.3-2, P3.3-3, and
P3.3-4. Isolate P3.3-1 was grouped with isolate P3.1-3 which also had similar reactivity to gp90 specific MCAbs. Although virus isolates could be distinguished antigenically on the basis of their immunoreactivity patterns, similar patterns of immunoreactivity were seen in isolates from animals infected in parallel. For example, isolates P3.1-3 and P3.3-1 recovered from the third and first febrile episodes of two animals infected in parallel exhibited similar pattern. Distinct changes were detected in specific epitopes of EIAV isolates recovered from a disease episode separated by only 2 weeks in a single animal, as shown by the differences in immunoreactivity pattern of isolates P3.2-1 and P3.2-2 or P3.3-1 and P3.3-2. This evolution of EIAV variants is very rapid. Detailed serological comparisons of the viral glycoproteins of EIAV isolates recovered from animals infected with the same inoculum (P3.1-1 through P3.1-4,
P3.2-1 through P3.2-5, and P3.3-1 through P3.3-4) showed also that they belong to different groups. Through serological comparisons we were able to identify 12 different antigenic variants among 16 virus strains (Table I I I .6).
Alterations of the glycoproteins involved both neutralization and non-neutralization epitopes during vivo passage (epitopes 88 Table III.6. Serotyping of 16 EIAV isolates on the basis of reaction with MCAbs to gp90 and gp45.
Antigenic variants Isolates included
Serotype1 Prototype
Serotype2 p3.2-3, p3.3-2,
Serotype3 p2.1-1
Serotype 4 p3.1-2
Serotype5 p3.1-1
Serotype 6 p3.2-2
Serotype 7 p3.1-3
Serotype 8 p3.3-1
Serotype 9 p3.1-4
Serotype 10 p3.2-4, p3.2-5
Serotype 11 p3.2-l
Serotype 12 p2.1-6, p3.3-4 89
90-B, 90-C, 90-D, 90-E, and 90-F). Epitopes 90-E and 90-F where found only in few isolates. In addition some neutralization epitopes conserved their configurational integrity during in vivo passage, for instance epitope 90-E is found on the prototype and only isolates
P2.1-1, P3.1-2, P3.1-4, P3.2-4, and P3.2-5. In a limited study, the four neutralizing MCAbs (82-1C2, 98-1D1, 115-3D7, and 128-2B9) were also tested in neutralization assays against EIAV isolates P2.1-1,
P2.1-6, P3.1-1, P3.1-2, P3.1-3, and P3.1-4, which displayed different immunoreactivity patterns with the neutralizing MCAbs in Western blot analysis (Fig. III.13). The virus dilution-constant ascitic fluid method was used since not all the isolates reproducibly produced CPE in cell culture (Issel, unpublished results). The results indicated that MCAbs 82-1C2, 98-1D1, 115-3D7, and 128-2B9 had a neutralization index of 2, 2.5, 2.5, and 2, respectively, against the prototype
EIAV. However, none of the isolates tested were neutralized by any of the four neutralizing MCAbs.
Ill.D . DISCUSSION
Lentiviruses derive their name from the slow time course of the infections they cause in humans and animals. The persistence and spread of these viruses despite host defenses and the origins and slow evolution of the diseases they cause pose fascinating problems in pathogenesis and vaccine development. Recently, these mechanisms assumed new significance with the addition of human immunodeficiency virus (HIV), the causative agent of AIDS, to the lentiviruses.
Several questions were addressed repeatedly to answer the enigmas of pathogenesis of lentivirus induced diseases. These include (1) How does a virus escape from the immunological surveillance mechanisms of 90 its host over a long period of time? (2) How is a virus disseminated in the animal in the face of competent immunological defense mechanisms? (3) How does the virus cause the pathological lesions? and, (4) Why does the infectious process evolve so slowly?
The emergence of mutant viruses with altered envelope glycoproteins, i.e., antigenic variation, is the postulated mechanism by which EIAV temporarily escapes immunological inactivation and persists in the horse. The present investigation has enabled us to determine the following factors which may assist in understanding antigenic variation of EIAV:
1) The identification and dissection of antigenic sites of the two main surface glycoproteins of EIAV.
2) The documentation of rapid epitope alterations in the virion surface glycoproteins of EIAV during persistent infections.
3) The pattern and extent of serological cross-reactions of
EIAV isolates, and an estimation of the minimum number of possible serotypes of EIAV.
4) The localization of determinant(s) involved in stimulating neutralizing MCAbs.
The development by Kohler and Milstein (65, 66) of a method for the isolation of immunoglobulin producing hybrid cells has proven to be useful to analyze proteins. Therefore, we were engaged in the production of MCAbs to study antigenic variation of EIAV and delineate the changes in the virion surface glycoproteins. We have described the production, characterization, and use of hybridoma antibodies specific for EIAV glycoproteins (gp90 and gp45). Our results showed that subtle antigenic differences among EIAV isolates exist. These were evident by definitive antigenic alterations at a
single or group of epitopes in the surface glycoproteins of EIAV in
persistently infected animals. The majority of isolates tested could
be distinguished antigenically on the basis of their pattern of
reactivity with the panel of MCAbs. This suggests a high degree of
heterogeneity of these isolates although a highly conserved epitopes
(epitopes 90-A and 45-A) among these strains was identified. The
genotypic, phenotypic and antigenic comparisons of EIAV isolates
revealed differences, but there were cases when two isolates appeared
identical by one method of analysis but differed by others. For
example, isolates P3.2-4 and P3.2-5 that appeared to be closely
related and could not be distinguished by our battery of MCAbs had
totally different patterns of oligonucleotide, peptide, and
glycopeptide maps (122, 129). In contrast, isolates P3.2-1 and
P3.2-4, which could not be distinguished by oligonucleotide or glycopeptide mapping of gp45, were found to be different by at least one epitope on gp90 (epitope 90-E). This correlated with data documenting differences in the gp90 peptide and glycopeptide maps of
these same isolates (122). Isolates P3.2-2 and P3.2-3, which could
not be distinguished by peptide maps, were distinguished by oligonucleotide and glycopeptide maps and by immunoreactivity with our panel of MCAbs. The dissim ilarity observed for these isolates with respect to glycopeptide and immunoreactivity pattern may
indicate the importance of glycosylation for the antigenicity of this particular epitope. Isolates P3.1-1, P3.1-2, P3.1-4, P3.1-3 and
P3.2-5 whose glycopeptide maps have proven sim ilar, could be distinguished by their immunoreactivity pattern with the panel of MCAbs. This may indicate that glycosylation may not be important for
the reactivity of this epitope. Previous studies showed that
peptides number 49 and 52 of gp90 (122) were present in one isolate,
absent from the next, and subsequently reappeared in later isolate.
These peptides may be correlated with several gp90 epitopes detected
by our panel of MCAbs that were reported to fluctuate among isolates
(epitopes 90-C, 90-D, and 90-E). In the same study 50% of all gp90
peptides were reported to be present in all isolates. These
conserved peptides may be present in epitope 90-A of gp90 noted in
all isolates. These observations document the heterogeneity of all
isolates tested and argue against the existence of two identical
isolates. All 16 isolates used for this study could be distinguished
by the combination of oligonucleotide, peptide, and glycopeptide maps or immunoreactivity pattern with MCAbs.
The availability of these MCAbs and the use of Western blot
immunoassay have allowed us to study the pattern and extent of
serological cross-reactions among EIAV isolates since some of these
MCAbs were capable of distinguishing between different EIAV isolates and could detect subtle antigenic differences among EIAV variants. It
is likely that this panel of MCAbs along with others which are being characterized could provide the basis for a method of typing EIAV if a large number of isolates are included in the comparison. Adequate cataloging of the protective glycoproteins of different EIAV strains might serve as an important means in the search for a protective vaccine. Therefore, we attempted to classify EIAV isolates into serologically distinct groups based on their immunoreactivity pattern with the panel of MCAbs. Our studies demonstrated that some 93 determinants, e.g., epitopes 90-A and 45-A were preserved through the process of variation of EIAV in infected animals. In contrast, other antigenic determinants (epitopes 90-B, 90-C, 90-D, 90-E, and 90-F) were unique for the prototype and some variants.
In comparison studies with our panel of 18 MCAbs we observed that both neutralizing and non-neutralizing MCAbs could differentiate between EIAV isolates. Monoclonal antibodies that reacted with conserved epitopes on gp90 and gp45 failed to neutralize EIAV. On the other hand, the variable regions of the envelope glycoprotein 90 were often associated with neutralizing epitopes. In some cases, the
MCAb which could neutralize prototype did not bind to the epitope of isolates and neutralization did not occur. In other cases, the MCAb could still bind to the variant but did not decrease infectivity.
Thus, one could conclude that antibody binding to a potential neutralization epitope and neutralization may be unrelated phenomena.
Identification and itemization of different neutralization epitopes simultaneously present on the virion is of particular interest for the study of the modification of neutralization pattern of all isolates after their passage in ponies. The possible coexistence of many different epitopes on a single virion was noted for the prototype and other isolates. Our present observations confirmed the presence of more than one distinct neutralization epitope on the virus. With our battery of neutralizing MCAbs, a minimum of three neutralization epitopes simultaneously present on the prototype EIAV were detected (epitopes 90-C, 90-D, and 90-E). However, using an additive ELISA our data suggest that at least 2 distinct neutralization epitopes were localized on gp90 (chapter IV). 94
Our comparison study revealed the fate of the neutralization epitopes and demonstrated that they may be lost during multiplication in vivo. However, upon in vivo virus replication an epitope may lose its function in virus neutralization but may keep its antigenic conformation unaltered. For instance epitope 90-E was found on the prototype, P2.1-1, P3.1-2, P3.1-4, P3.2-4, and P3.2-5 but was not detected on other isolates tested. Similarly, epitopes 90-C and 90-D were found on some but not all virus isolates. Virus isolates that kept the antigenic conformation of the neutralization epitope unaltered carry 2 of the 3 neutralization epitopes found on the prototype. However, isolates P2.1-1 and P3.1-2 have all 3 epitopes
(90-C, 90-D, and 90-E).
In the understanding of antigenic variation of EIAV during multiplication in vivo the most pertinent finding was that the same epitope can have alternate functions in the neutralization of virus infectivity. By performing cross neutralization with MCAbs and
Western blotting with MCAbs in parallel we demonstrated that, due to mutation, an epitope can lose its function in viral neutralization without modifying its antigenic specificity. Isolates from individual animals, e.g., isolates (P2.1-1, P2.1-6), (P3.1-1 through
P3.1-4) and (P3.3-1 through P3.3-4) proved to be distinct by cross-neutralization studies using homologous serum (117, Rwambo et a l., submitted). Although neutralizing epitopes (90-C and 90-D) were present for example on both isolate P3.1-1 and P3.1-2 which have proven to be unique by a cross-neutralization te st, they may not function as neutralizing epitopes in these isolates. This is one example where binding and neutralization of epitopes are discrete functions. However, glycosylation may play a role in this case by masking epitopes on the protein backbone that may otherwise be potential targets for neutralization. A more comprehensive picture of epitope variation was obtained by analyzing isolates from animals infected in parallel. Based on these results, we concluded that a mutation can affect a neutralization site in two ways: either by modifying its antigenic configuration or by negating its function in virus neutralization. It seems then, that during selection of one particular isolate from its progenitor, neutralization epitopes (for example epitope 90-E on P3.1-1) were lost by conformational alterations. Other epitopes retained their conformation (epitopes
90-C and 90-D on P3.1-1) but lost their function in virus neutralization. Despite the significant progress made in the study of the mechanism of virus neutralization, no explanation is available for the mechanism by which a virus epitope can conserve its antigenic configuration while losing its function in neutralization. The high degree of variability is well illustrated by the complete conversion of the epitope pattern on some isolates (epitopes 90-C, 90-D, and
90-E on P3.1-2) to that of the prototype. These observations suggest that mutations involving residues outside the antigenic site would be expected to exert considerable effect and may cause conformational readjustment of the whole antigen. Recently however, antigenic variants of type 1 poliovirus selected in the presence of neutralizing MCAb had mutations situated in VP3, outside of antibody-binding site which is present in residues 93 through 103 of
VP1 (7). Genome sequencing of such antigenic variants revealed the mutation to always be located outside of the MCAb binding site. 96
These observations may have relevance to EIAV infections. Since
EIAV causes a persistent viremia and from our observation it induces
the production of non-neutralizing antibodies, then infectious
immune complexes may have important role in the pathogenesis of EIA.
Infectious immune complexes have been demonstrated in EIA (96). In
addition, antibodies may combine with free viral antigens
(glycoproteins, proteins) that may be shed from the virions. The
presence of non-neutral ized virions as well as antigen immune
complexes in the serum may produce immune complex disease. These in
turn may block immune cytolysis of virus-infected target cells by T
cells or complement fixing antibodies. These non-neutralizing
antibodies may function by saturating all possible antigenic
determinants on EIAV or may cause simple aggregation of EIA virions which may not result in intrinsic neutralization. On the other hand,
they may combine with EIAV in less than saturation concentrations, yet protect the virus against additional binding of antibodies.
This last attribute may involve binding of interfering antibodies to
non-critical sites that prevent binding of potential neutralizing
antibodies.
It is interesting to note that only two immunoreactivity patterns were seen for gp45 specific MCAbs. Seven gp45-specific MCAbs
recognized a highly conserved epitope (epitope 45-A) on all 16 EIAV
isolates, while one gp45-specific MCAb recognized an epitope (epitope
45-B) on selected isolates. Data provided by additivity ELISA
(Chapter IV) indicated that all 7 MCAbs may recognize the same epitope on gp45. It appears from our studies that both gp90 and gp45 may be involved in the antigenic changes observed in EIAV variants. Although not studied here, no structural variations were observed in the viral core proteins p9, pl5, and p26 of several viral isolates as tested by peptide mapping (102, 129). Both of these observations indicate that antigenic variation in EIAV during infection may be restricted to the surface polypeptides (gp90 and gp45) which are subject to host immune selection pressure. However, our results showed that gp90 appears to contribute more to this variation than does gp45. In addition the results indicate that although gp45 is strongly recognized by the host (as seen from
Western blot immunoassay using horse immune serum) its contribution to the observed antigenic variation in EIAV appears to be less than gp90. Heterogeneity of gp90 was documented by identification of nine different serologic groups. In contrast, only two serologic groups of gp45 were identified, suggesting a more moderate degree of conservation. In some cases viruses having similar immunoreactivity patterns with gp90 specific MCAbs differed with respect to gp45
(Groups 3, 4 and 7). Other isolates, e.g., all those from P3.1, had identical immunoreactivity patterns with respect to gp45, but differed with respect to gp90 immunoreactivity. These observed alterations in EIAV glycoproteins point out to the complexity of the situation for EIAV variation. Alterations observed in gp45 antigenicity may interact with gp90 thereby affecting overall display of antigenicity of very important epitopes. If this is the case, it represents a complex interaction, yet unique which may require further investigation.
The results confirmed that antigenic variation can occur in
EIAV isolates recovered from disease episodes separated by only 14 to 98
16 days (P3.2-1 and P3.2-2; P3.3-1 and P3.3-2). Thus, the time required for apparent antigenic variation in EIAV is considerably shorter than that reported for visna virus, which requires one to two years for the emergence of antigenic variants in the persistently infected animal (110). The spectrum of EIAV variants which arose in the three animals infected in parallel were distinguished antigenically on the basis of their immunoreactivity pattern. Each isolate in the same animal is different antigenically from the preceeding virus. However, the emergence of these variants appears to be random as no predictable pattern of immunoreactivity was observed in the three animals infected in parallel with the same virus inoculum. These findings differ from those with visna virus, for which a similar pattern of variants evolved in parallel persistent infections (23). Our findings paralleled those of AIDS virus (HIV), in which genetic variants appear to have evolved randomly (49). In EIAV, although viruses isolated from a given animal were different antigenically from each other, they appeared to have similar immunoreactivity to viruses isolated from other animals infected with the same inoculum. This is in contrast to the findings in HIV where genetic variants from any one patient were observed to be more related to each other than to viruses from other individuals
(49). However, HIV isolates that appeared to be related toeach other may actually represent distinct viral forms that have persisted in individuals for a long time. These individuals may have been exposed to more than one genotypic form of HIV which may account for the difference in these observations, while ponies experimentally infected received the same virus inoculum. Although EIAV has the potential to change dramatically in each infected horse, our data suggest that only a finite number of changes are possible. Through serological comparisons we were able to identify 12 different antigenic variants among 16 virus strains. Also, the results have shown that the sequential isolates differ in their surface glycoproteins, and that these structural changes, as well as changes in viral RNA, were paralleled by changes in antigenicity as monitored by neutralization, cross-neturalization, and immunoreactivity pattern using a panel of MCAbs. The data support the hypothesis of antigenic variation which involves frequent point mutations in the envelope gene, which encodes the virion envelope glycoproteins. When alterations of antigenic determinants of these surface proteins occur, the emerging variants are not affected by immune pressures effective against previous variants. This new variant then multiplies in an unrestricted fashion until immune responses to the novel variant effectively check virus replication. As this cycle repeats itself at frequent intervals, it results in the chronic
(cyclic) form of clinical disease most recognized as EIA. It is important to mention here that the 15 virus isolates obtained from plasma taken at distinct febrile episodes had been purified by end point dilution before expansion and propagation to isolate the predominant virus strain (100). Thus our documentation Of rapid epitope alteration in the virion surface glycoproteins among sequential virus isolates demonstrated that each disease episode is associated with a unique virus strain. Thus antigenic variation is an important mechanism for the persistence and dissemination of EIAV and seems to be a more persuasive mechanism than what has been 100
reported for visna virus. In visna, variants do not replace the
infecting serotype via antibody selection mechanisms and, in most
long-term infections, the initial infecting virus strain persists without the emergence of antigenic variants. These observations argue against the role of antigenic variation as an important means of dissemination of visna virus (86, 141).
Persistence of EIAV may also involve other mechanisms in addition to antigenic variation. This is not surprising in view of the complex life cycle of retroviruses. Being a retrovirus, EIAV is provided with an indefinite survival mechanism because of its ability to integrate proviral DNA into host DNA. Therefore, a latent infection may be established, and no mechanisms exist for the immune system to eliminate a latent viral genome. The continued presence of
EIAV in the blood stream may involve predominantly, if not exclusively, infected monocytes (19). Therefore, it is possible that these mobile cells conceal the virus genome and convey i t without detection to other sites. This may be accomplished by a restricted level of viral RNA synthesis in the infected cells, thereby limiting viral release. Biologically significant mutations may also result from errors of reverse transcription by the mistake-prone viral trancriptase (151).
Monoclonal antibodies which react with a single antigenic determinant are the most specific biological probes available to discern the biological and structural properties of viral proteins.
Their use to define epitopes important in stimulation of neutralizing antibody and their relevance in constructing an effective vaccine for
EIAV are incontrovertible. Our results showed that gp90 possesses determinant(s) very important in neutralization (epitopes 90-C, 90-D, and 90-E). The studies with MCAbs also suggest that gp90 is the major glycoprotein available at the surface of the virus which acts as a target for neutralizing antibodies. This may indicate that critical epitopes important for neutralization are clustered on gp90.
On the other hand, gp45, a hydrophobic transmembrane protein, may play a more important structural role. However, polyclonal sera which also neutralize EIAV infectivity contain antibodies to multiple epitopes including those found on gp45. These may play a role in protection which is yet to be determined. Immune factors to gp45 may participate in "cooperative neutralization" a synergistic phenomena which results when two antibodies cause a greater loss of infectivity than the sum of their effects individually (33). Alternatively gp45 may carry determinants that interact with specifically sensitized cytotoxic T cells.
Neutralization studies indicated that the nature of the critical epitopes for induction of neutralizing antibodies in mice may not be affected by treatment with deoxycholate since mice immunized with dissociated virus produced neutralizing antibodies. The epitopes of gp90 which reacted with neutralizing MCAbs (82-1C2, 98-1D1, 115-3D7, and 128-2B9) appear to be continuous since they resisted sodium dodecyl sulfate treatment. Their neutralizing effect was demonstrated to be independent of complement source. Additionally complement seems to have no effect on the neutralization capability of immune horse serum. Monoclonal antibodies that reacted with conserved epitopes on gp90 or gp45 failed to neutralize EIAV. On the other hand, our results showed that the variable regions of gp90 often contained neutralizing epitopes. Three antigenic sites
(epitopes 90-C, 90-D, and 90-E) functional in neutralization could be
tentatively identified based on the immunoreactivity pattern of gp90.
However, data provided from topographical mapping showed at least two distinct epitopes important for neutralization (Chapter IV). These studies provided information toward identifying the immunoreactive proteins of EIAV through the use of MCAbs. Analyzing the epitopes on viral proteins represents a very important tool for identifying critical site(s) involved in virus neutralization. However, our study has shown that epitopes on gp90 which are variable and highly antigenic may represent type-specific epitopes that react with neutralizing MCAbs. In this study, none of the isolates tested were neutralized by any of the 4 neutralizing MCAbs. It would be of interest to check whether other MCAbs that neutralized homologous virus could neutralize heterologous EIAV isolates in cross-neutralization tests. If cross neutralization occurs, then a common determinant might be identified and serve as a basis for a subunit vaccine developments. This would be an ideal vaccine, but alternative vaccines could be constructed by incorporating immunogenic fragments of different epidemiologically significant variants. Our studies showed some cross-reactivity of isolates from different animals, suggesting that there may be a finite number of changes. In addition the frequency and severity of clinical episodes of EIA decrease with time in most infected horses and they often become asymptomatic carriers (60). This cessation of clinical illness is probably achieved by the ability of infected animals to mount immune responses to a wide enough diversity of EIAV strains to 103 finally achieve protective immunity. In this case immune responses to the virus continuously check virus replication, but the virus is not eliminated from infected horses. Both of these observations indicate that the possibility of the existence of a common antigenic determinant is high, and its identification will undoubtedly lead to the production of a group specific immunogen. Extensive evidence does, however, warrant caution when a live attenuated vaccine is contemplated for EIA. This is mainly because the carrier state would be established and the virulence would undoubtedly be restored by passage of virus in ponies even after attenuation from long term passage in tissue culture (116, 117).
With this discussion in mind, I would like to outline the requirements regarding an EIAV vaccine which would have to be fulfilled in order for the vaccine to be acceptable. Such a vaccine would have to be noninfectious, prevent persistent infection, not stimulate antibody which might interfere with the serological test commonly used to detect EIAV infected horses, and confer protection against all possible antigenic variants of EIAV. Finally, effective control of EIA requires the identification of cells and tissues in the host where the virus is sequestered after the initial immune response. Identifying these cells and tissues in which the virus seeks refuge from the immune defenses and which provide the virus with a potential sanctuary may help us to further understand the pathogenesis of the disease and develop more effective procedures for its control. CHAPTER IV
ANTIGENIC MAPPING OF THE ENVELOPE PROTEINS OF EQUINE INFECTIOUS
ANEMIA VIRUS: IDENTIFICATION OF A NEUTRALIZATION DOMAIN AND A
CONSERVED REGION ON GLYCOPROTEIN 90
IV.A. INTRODUCTION
Equine infectious anemia virus, a member of the family
Retroviridae, subfamily Lentivirinae, causes a naturally occurring disease that is of great economic importance to the horse industry
(60). The virus contains two surface glycoproteins (gp90 and gp45) and four major nonglycosylated internal proteins referred to as p26, pl5, p ll, and p9 (100, 120).
Antigenic variation occurs in EIAV and is thought to involve frequent point mutations of the env gene, which codes for viral surface proteins (61, 104). Genetic and antigenic variants of EIAV have been previously demonstrated by oligonucleotide and peptide mapping and most recently by monoclonal antibody analysis (54, 102,
121, 122, 129). The alterations, which seem to be confined to the envelope glycoproteins, may be of great importance for the biology and pathogenicity of EIAV and may allow the emerging variant to escape from neutralization by preexisting antibody. Data from chronically infected horses suggest that production of broadly reactive immune responses to EIAV might eventually occur, which could result in effective group-specific immunity. We previously described a panel of MCAbs recognizing conserved and non-conserved epitopes on both gp90 and gp45 of EIAV. Sites that are both prone to variation and highly antigenic may represent type-specific epitopes of gp90 that react with neutralizing MCAbs (54). In the present report, m 105 selected MCAbs were used to map epitopes of EIAV glycoproteins. The procedures employed demonstrated the presence of two distinct epitopes on gp90 that are important in neutralization. These epitopes may be of major importance for the preparation of a synthetic vaccine for EIA since competitive ELISA demonstrated the presence of a common neutralization domain on gp90. Therefore, this particular domain may be of some biological significance in the induction of protective immunity. In addition, fragmentation studies of gp90 revealed an immunoreactive fragment(s) of 12K which carries an epitope shared by various serologically distinct EIAV isolates.
IV.B. MATERIALS AND METHODS
IV.B.l. Viruses
The prototype EIAV used in this study was obtained by propagation in fetal equine kidney (FEK) cells of the cell-adapted
Wyoming strain of EIAV (88). All virus isolates were produced by serial passage of the prototype strain of EIAV in Shetland ponies as described previously (54, 116, 117, 122, 129). The prototype strain and virus isolates were propagated in FEK cells and were purified as described in detail previously (100).
IV.B.2. Glycoprotein purification
Procedures for EIAV glycoprotein purification by lectin affinity chromatography have been described in detail previously (101).
IV.B.3. Production of monoclonal antibodies
The production and characterization of the murine MCAbs to EIAV glycoproteins used in this study has been previously described (54,
Chapter II and III). Briefly, 10 MCAbs were specific to gp90 and 8 to gp45 of EIAV. Four of these MCAbs to gp90 neutralized infectivity 106 of prototype EIAV, and 6 potential epitopes of gp90 and 2 for gp45 were suggested on the basis of immunoreactivity with a battery of
EIAV strains (Fig. IV.1).
IV.B.4. SDS-PAGE and Western blot assay
Procedures for the analysis of EIAV proteins by SDS-PAGE and
Western blot with either MCAbs or reference EIA-positive horse serum have been described (54, 100, 129, Chapter II and III).
IV.B.5. Generation and detection of gp90 fragments
Fragments of gp90 were generated by limited digestion of either
intact EIAV (150 yg) or lentil-lectin purified glycoprotein fractions of EIAV (25 yg) with 10 yg papain (Sigma Chemical Co. St. Louis, M0)
in PBS for 2 h at 37°C (total volume 50 yl). The reaction was terminated by the addition of 40 yl 0.01 M phosphate buffer, pH 7.2,
10 yl 10% SDS, 1 yl 2-mercaptoethanol and boiling for 3 min.
Proteolytic fragments were then separated on SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted with either MCAbs or reference EIA-positive horse serum to identify the immunoreactive fragments. Western blots were modified when using MCAbs to include the addition of goat anti-mouse immunoglobulin G (IgG) prior to the 1 PR addition of I-labeled protein A (13). The immunoreactive fragments were identified by autoradiography using Kodak X-Omat film (XRP-5).
Glycosylated peptide fragments were stained using concanavalin A and peroxidase as described by Clegg (21).
IV.B.6. Enzyme-linked immunosorbent assay - additivity test
Purified EIAV virions were disrupted with 0.5% deoxycholate in
Tris-saline buffer (0.02 M/0.1 M), diluted in TEN buffer (Tris [0.05
M], EDTA [0.001 M], NaCl [0.15 M]), dispensed at 300 ng/well, and 107
P3.1-1 P3.2-3 P3.2-2 P 3.3-2
P3.3-1
prototype
P3.2-1
P3.1-3 P 3.3-4 P 3.2-4 P 3.2-5
FIGURE IV.1. Operational antigenic map of 16 EIAV isolates. 108 allowed to adsorb to a flat bottom Immulon 1 microtiter plate
(Dynatech, Torrance, CA) for 16 to 20 h at 37°C incubation. The antigen was fixed for 5 min with paraformaldehyde (4% W/V, pH 7.2) and blocked with 5% BSA for 2 h at room temperature. Epitope specificity was analyzed using concentrations of antigen and ascitic fluids shown to produce 50% saturation under the reaction conditions employed (see Results). After 3 washings with TEN buffer, 50 yl of the B-D-galactosidase-labeled goat anti-mouse immunoglobulin
(Southern Biotechnology Associates, Birmingham, AL) were added to each well at a dilution of 1:250, and incubated for 1 h at room temperature. After 3 further washings, 100 yl of assay buffer containing 0.1% 0-nitrophenyl-$-D-galactoside (0NPG, Sigma Chemical
Co., St. Louis, M0), 10 mM Tris base, io mM sodium chloride, 10 mM magnesium chloride, and 10 mM 2-mercaptoethanol, pH 7.5, were added to each well. After incubation for 30 min at room temperature, 50 yl of 1.43 M sodium carbonate were added to each well to stop the enzymatic reaction. Absorption at 415 nm was recorded using an automated ELISA plate reader (Dynatech). A bright yellow reaction product formed in wells retaining enzyme. Each dilution of MCAb was run in triplicate and each experiment was repeated at least twice.
The results of the additive ELISA were expressed quantitatively as an "additivity index" (AI). The AI has been defined (40) for a pair of antibodies as:
AI = 100 ([(2A1+2)/(A1 + A2)] - 1), where Ap A2 and A^+2 represent the optical densities obtained with the first antibody alone, the second antibody alone, and the two antibodies together, respectively. An AI of 0-50% was considered not additive, while an 109
AI of £50% was considered additive.
IV.B.7. Competition binding studies
These studies compared the ability of unlabeled polyclonal horse sera to compete with neutralizing MCAbs in binding to EIAV determinants as measured by ELISA. Density gradient-purified virions were disrupted with 0.5% deoxycholate in Tris-saline buffer and maintained at 200 yg/ml concentration. This virus preparation was diluted in TEN buffer to a concentration of 10 yg protein/ml, dispensed at 0.5 yg/well, and incubated for 16-20 h at 37°C in
96-well Immulon 1 m icrotiter plates. The antigen was fixed as described for the additive ELISA and plates were blocked with 3% BSA for 2 h at room temperature. Plates were washed 3 times with
TEN/0.05% Tween 20 buffer prior to use. Serial two-fold dilutions of unlabeled polyclonal sera (reference EIA-positive horse serum) were performed using TEN buffer as a diluent and adding 50 yl to each well. In addition, EIA-negative horse sera as well as sera collected early (35 days post-infection) and late (203 days post-infection) in the course of an experimental infection of a pony (#F135) with EIAV were included in the test (Rwambo et al., submitted). After a 30 min incubation at room temperature, 50 yl of MCAb diluted in TEN buffer was added. The mixture of polyclonal sera and MCAb was incubated for
1 h at room temperature. The plate was washed as before, and then peroxidase-labeled goat anti-mouse immunoglobulins (Kirkegaard and
Perry Laboratories Inc., Gaithersburg, MD) was added (50 yl per well) at a 1:500 dilution in TEN and incubated for 1 h at room temp. The plates were washed as before and developed using the substrate ABTS
I (2, 2 -azino-bis(3-ethylbenz-thiazoline-6-sulfonic acid), diammonium 110
salt (Sigma). The reaction was allowed to develop for 20 min at room
temperature, and the optical density (O.D.) at 490 nm was recorded
with a spectrophotometer (Dynatech Laboratories). Blank wells
(EIAV-coated wells exposed to all reagents except MCAb) were included
in each row of the plate. Viral antigen and MCAbs were pretested and
diluted to give an approximate optical density of 0.2-1.0 units in
the absence of competition.
IV.C. RESULTS
IV.C.l. Epitope mapping
Monoclonal antibodies specific for gp90 and gp45 were used in an
additive ELISA to determine their epitope specificities. The test was
based on an estimation of the number of epitopes simultaneously
available to a pair of antibodies on the antigen. A key requirement
for this test is that the antigen coated on wells of the ELISA plate
be saturated with each individual antibody to be tested alone. The
dilution of antibody which provides saturation would then be used in
combinations with other antibodies to determine the results of
simultaneous binding of the antibodies in the presence of each other.
Therefore, titratio n s of MCAbs were performed to determine saturation
curves of the EIAV antigen by the B-D-galactosidase ELISA. The concentration of EIAV antigen and each MCAb were standardized so that
the optical density of the colorimetric reaction was approximately one half the possible maximum.
The results obtained with four neutralizing MCAbs tested
indicated that dilutions of 1:256, 1:128, 1:256, and 1:64 of MCAbs
82-1C2, 98-1D1, 115-3D7, and 128-2B9, respectively saturated the epitopes which each of these individual MCAbs bound separately (Fig. I l l
IV.2). The additivity ELISA for the pairs 82-1C2 and 128-2B9, 98-1D1 and 128-2B9, or 115-3D7 and 128-2B9 gave AIs of 84%, 72%, and 76%, respectively (Table IV.I). In each case, these specified pairs of
MCAbs appeared to simultaneously bind gp90 , i .e ., they reacted with distinct, non-overlapping epitopes of gp90. In contrast, the optical density recorded for 82-1C2 and 98-1D1, 82-1C2 and 115-3D7, or 98-1D1 and 115-3D7 was nearly equal to that reached for each MCAb separately
(AI values of 11%, 9%, and 15%, respectively). This suggests that in each case, the two MCAbs competed for the same or overlapping antigenic sites of gp90.
Using the additivity te st and index we extended our comparison of the epitope binding specificities of non-neutralizing MCAbs specific for gp90. An AI of 53% was observed with MCAbs 86-1E3 and
128-2B9, while no AI value indicative of additivity was seen with all other MCAbs pairs tested (Table IV.2). This suggests that the binding of MCAbs 86-1E3 (epitope 90-A) and 128-2B9 (epitope 90-E) was additive, and thus distinct epitopes were recognized. All other pairs of non-neutralizing anti-gp90 MCAbs were not additive, suggesting the recognition of similar, contiguous, or overlapping regions on gp90. Similarly, results obtained with eight MCAbs specific for gp45 clearly showed that the binding of MCAb 90-1C1and each of the other seven MCAbs: 75-1F2, 92-1E6, 101-2F10, 109-1A6,
117-1C5, 120-1H9, and 105-3C8 was additive (AI values of 63%, 68%,
70%, 60%, 63%, 58%, and 84%, respectively) (Table IV.3). These results suggested that only MCAb 90-1C1 was recognizing an epitope on gp45 distinct from that one bound by the other seven. In contrast, the binding of all other pairs of gp45-specific MCAbs was not etaiig oolnl niois 8-C, 1-D, 811 and 98-1D1 115-3D7, 82-1C2, antibodies: monoclonal neutralizing F IGU RE IV.2. Saturation curves of the coated EIAV antigen with with antigen EIAV coated the of curves Saturation FIGURE IV.2. 128-2B9.
Absorbance at 415 nm .48 .64 .16 .80 .32 .28 .42 .16 .16 .64 .32 .48 .80 14 56 48 70 64 32 80 1 26 2 64 128 256 512 eirclo te iuin f neutralizing of dilution the of Reciprocal monoclonal antibodies monoclonal 16 21 2 82-1C 115-3D7 128-2B9 98-1D1
16 112 Table IV.1. Additive enzyme-linked immunosorbent assay.
2 Monoclonal antibodies^- Optical density Expected value Additivity at 415 nm if additive index (%)
82-1C2 + 98-101 0.60 1.08 11
82-1C2 + 115-3D7 0.48 0.88 9
82-1C2 + 128-2B9 0.60 0.65 84
98-1D1 + 115-3D7 0.59 1.02 15
98-1D1 + 128-2B9 0.68 0.79 72
115-307 + 128-2B9 0.52 0.59 76
Each monoclonal antibody was tested separately and an optical density of 0.47, 0.61, 0.41, and 0.18 was obtained at saturating dilution for MCAb 82-1C2 (1:256 dilution), 98-1D1 (1:128 dilution), 115-3D7 (1:256 dilution), and 128-2B9 (1:64 dilution) respectively.
2 Expected value is the sum of the optical densities obtained for each ascitic fluid alone. Table IV.2. Additivity index of gp90 specific monoclonal antibodies
Monoclonal antibodies 86-1E3 87-1E7 82-1C2 115-3D7 128-2B9 85-1E11
86-1E3 ------16 9 12 53 13
87-1E7 16 ------11 5 9 14
82-1C2 9 11 ------9 84 15
115-3D7 12 5 9 ------76 17
128-2B9 53 9 84 76 ------14
85-1E11 13 14 15 17 14 ------
Additivity index is shown in % Table IV.3. Additivity index of gp45 specific monoclonal antibodies
Monoclonal antibodies 75-1F2 92-1E6 101-2F10 109-1A6 117-1C5 120-1H9 105-3C8 90-1C1
75-1F2 ------15 7 14 20 29 22 63
92-1E6 15 ------7 16 23 7 13 68
101-2F10 7 7 ------11 14 16 9 70
109-1A6 14 16 11 ------9 12 16 60
117-1C5 20 23 14 9 ------11 18 63
120-1H9 29 7 16 12 11 ------17 58
105-3C8 22 13 9 16 18 17 ------84
90-1C1 63 68 70 60 63 58 84 ------
Additivity index is shown in %. 116 additive, suggesting that in each case the two MCAbs competed for the same or overlapping antigenic site of gp45.
IV.C.2. Competitive ELISA
Four neutralizing MCAbs, specific to gp90, were studied by competitive ELISA to see whether epitopes recognized by murine MCAbs were the same or different from those recognized by polyclonal sera from EIAV-infected horses. The results of the four neutralizing
MCAbs are presented in Fig. IV.3, and the summary of all assays is presented in Table IV.4.. Polyclonal sera collected late in the infection as well as reference EIA-positive horse serum (both with neutralization activity) were found to compete effectively with all four neutralizing MCAbs. This competition can be interpreted to indicate either similar epitope(s) recognition or that these epitopes have close spatial relationships and constitute a single large neutralization domain on gp90 of EIAV. Serum collected early in the infection, i.e., 35 days after infection, was unable to neutralize
EIAV and failed to cause a significant inhibition of binding of any of the four MCAbs. The early serum, however, was reactive with EIAV glycoproteins gp90 and gp45 in both ELISA and Western blot (Fig.
IV.4). A 1:256 dilution of the reference EIA-positive horse serum, on the other hand, neutralized 100 TCID^q of EIAV in a microtiter assay. Normal horse serum, negative for EIA as tested by ELISA,
Western blot, agar gel immunodiffusion, and neutralization tests, failed to inhibit binding of any of the MCAbs to virus glycoproteins.
IV.C.3. Analysis of qp90 fragments
Limited digestions of the glycoproteins of the prototype EIAV as 117
100 90 80 70 g 60 55 50 5 40
20
4 8 16 32 64 128 256 512 10242 RECIPROCAL ANTIBODY DILUTION
FIGURE IV.3. Competitive inhibition of neutralizing MCAbs 82-1C2, 98-1D1, 115-3D7, and 128-2B9 binding to EIAV glycoprotein in solid phase-ELISA. Serial two-fold dilutions of polyclonal sera (reference EIA-positive horse serum) were tested for their ability to inhibit the binding of neutralizing MCAbs as described in the text. Symbols: (□) 82-1C2, ( O ) 98-1D1, ( 0 ) 115-3D7, ( A ) 128-2B9. Each data point represents the mean of two replicate assay determinations. 118
TABLE IV.4. Summary of competitive ELISA
Neutralizing monoclonal antibodies EIA-positive horse sera Natural Experimental Clone number Reference Early Late
82-1C2 + + 98-1D1 + + 115-3D7 + + 128-2B9 + +
^Reference: EIA-positive serum from a horse naturally infected with
EIAV. Experimental: serum collected 35 days (early) and 203 days (late) post-infection from a pony experimentally infected with EIAV. + indicates >50% inhibition by the unlabeled polyclonal sera (competitor) 1 1 9
1 2 3
gp 45
p26
FIGURE IV.4. Western blot analysis of EIAV. Proteins of density-gradient purified EIAV were separated by SDS-PAGE, transferred to nitrocellulose membrane, and immunoblotted with both natural and experimental EIA-positive horse sera. Lane 1 and 2, serum collected 35 days (early) and 203 days (late) post-infection from a pony experimentally infected with EIAV. Lane 3, EIA-positive serum from a naturally infected horse used as a reference. 120 well as whole virus were tested with reference EIA-positive horse serum and with both neutralizing and non-neutralizing MCAbs against gp90 to identify immunoreactive fragment(s). All neutralizing MCAbs
(82-1C2, 98-1D1, 115-3D7 and 128-2B9) and MCAb 86-1E3 showed reactivity with two fragments of 12K and 20K. Papain digestion for more than 2 h removed intact gp90 from whole virus and generated immunoreactive peptide(s) (Fig. IV.5). The size of the reactive fragment(s) was not further reduced with overnight digestion.
Western blots of identical digests with the reference EIA-positive horse serum also indicated reactivity with 12K as well as a 20K fragments (data not shown). The 20K fragment appeared to be an intermediate cleavage product that was also evident in Western blots with MCAbs. Glycoprotein staining of these blots indicated that the larger fragment (20K) was clearly glycosylated while the smaller one
(12K) was not.
The prototype EIAV and a panel of serologically distinct isolates (P3.2-1, P3.2-2, P3.2-3, P3.2-4, and P3.2-5) obtained from consecutive febrile episodes of experimentally inoculated ponies
(54, 116, 117, 129) were digested with papain and immunoblotted with
MCAbs to identify common immunoreactive fragments. The non-neutralizing MCAb 86-1E3 reacted with gp90 of all EIAV isolates tested in the undigested controls (Fig. IV.5, panel B). However, the epitope recognized by this MCAb was present on the 12K fragment generated from the same virus isolates (Fig. IV.5, panel A).
IV.D. DISCUSSION
Antigenic variation is a prominant feature of EIAV and has been documented previously (54, 102, 122, 129). Alterations in one or 121
1 2 3 4 5 61 2 3 4 5 6
gp9Q ■ ■ HP 67K
43 K
■
|L . 14K»
FIGURE IV.5. Western blot analysis of gp90 fragments of several serologically distinct isolates of EIAV obtained from consecutive febrile episodes. Fragments of gp90 of these viruses were generated by papain digestion, separated by SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with MCAb 86-1E3. Western blots of papain treated (panel A) and untreated (panel B) prototype EIAV (1), antigenic variants P3.2-1 (2), P3.2-2 (3), P3.2-3 (4), P3.2-4 (5), and P3.2-5 (6). more epitopes of the surface glycoproteins have been observed among sequential isolates of EIAV during persistent infection (54). In the present study, we dissected the antigenic sites of the surface glycoproteins of prototype EIAV. By our methods, i t was possible to compare the epitope specificities of ten MCAbs against gp90 and eight
MCAbs against gp45 of EIAV produced by independent clones. These analyses resulted in the identification of 6 antigenic epitopes, 4 on the exterior glycoprotein (gp90) and 2 on the transmembrane protein
(gp45). The combined results were the basis for the construction of a proposed model for the relationships and distributions of epitopes on glycoproteins of EIAV. According to the predicted topographical model, epitopes on gp90 were found to cluster in at least four antigenic sites (Fig. IV.6). Three neutralizing MCAbs (82-1C2,
98-1D1, and 115-3D7) reacted with a single topographic site termed
C/D. A second site (E) which appeared to be distinct from site C/D was characterized by one neutralizing MCAb 128-2B9. A third site (A) was characterized by three non-neutralizing MCAbs (86-1E3, 95-1G8, and 114-3A7) that were additive with neutralizing MCAb 128-2B9 (site
E). Site A appeared to be close to site C/D but distinct from site
E. The fourth site B/F was characterized by three non-neutralizing
MCAbs (87-1E7, 71-1A9, and 85-1E11) that were not additive with all other MCAbs tested. These observations may suggest a highly overlapping binding site.
Analysis of gp45-specific MCAbs showed that 7 out of 8 MCAbs recognized either the same, a contiguous, or an overlapping epitope.
Interestingly, previous studies indicated that all seven MCAbs have similar serological specificity (54), and therefore it is possible 123
FIGURE IV.6. Schematic model of the proposed topological arrangement of the epitopes on gp90 of EIAV. Hatched areas indicate overlap. that they recognize a highly conserved epitope even though these
MCAbs were derived from different fusions and independent hybridoma clones. The results of the additivity ELISA indicated that at least two distinct epitopes were recognized on gp45 which paralleled our earlier results (54) and demonstrate that a given serologically distinct specificity in this case was equivalent to a given distinct epitope. It should be emphasized that the panel of MCAbs used in this study was generated by chance but with the intention that they discriminate among EIAV variants. Epitope mapping of EIAV using our diverse panel of MCAbs revealed the presence of at least 2 epitopes
(one on gp90 and one on gp45) conserved in various strains of EIAV.
The conserved epitope localized on gp90 may represent peptide segments essential for protein structure or function, while other non-essential sites are more prone to variation and at the same time highly antigenic. Conservation of epitopes appears to be a prominent property of gp45, which may reflect the fact that much of this glycoprotein is sequestered by the viral lipid bilayer, as proposed for gp41 of the human immunodeficiency virus (43). Most importantly, however, conserved epitopes may represent valuable diagnostic antigens since both glycoproteins are intensely recognized by the reference EIA-positive horse serum and are detected earlier than precipitating antibodies in the course of an infection (Issel, unpublished results).
Results of the competitive ELISA demonstrated that EIA-positive horse serum contained gp90-specific antibodies directed against the neutralization domain of gp90. Polyclonal equine sera (reference and late serum) blocked the binding of all four neutralizing MCAbs, 125 indicating that the neutralization epitopes bound by these antibodies are localized presumably in an immunodominant domain on gp90. The block of neutralizing MCAbs was shown to be specific since serum collected early in the course of experimental infection (which reacted with EIAV glycoproteins in Western blot assay but had no neutralizing activity) did not inhibit the binding of all four neutralizing MCAbs. The neutralization domain of gp90 was apparently of sufficient size to incorporate more than one operationally distinct epitope.
Several conclusions could be drawn from our studies. Our studies reveal a continuum of epitopes which appear to consist of overlapping sites. Of the 10 gp90-specific MCAbs analyzed, a high degree of overlap was observed which suggests a cluster of epitopes.
Therefore, a large, single heterogeneous domain is apparent in our model of gp90, which includes four different epitopes based primarily on functional and serological studies. Two distinct
(non-overlapping) antigenic sites were shown to contain epitopes involved in neutralization, demonstrating that this function is not necessarily restricted to a single site on EIAV glycoproteins.
Therefore, different sites on gp90 could have the potential to be selected for the development of a synthetic vaccine for EIA.
Competitive ELISA indicated that certain sites on gp90 are immunogenic both in mice and horses, suggesting that these sites may represent common immunodominant neutralization domains with several interrelated epitopes on gp90. This site may have a significant biological importance in the induction of protective immunity. Some gp90-specific MCAbs that have no detectable neutralization capability 126 may block the binding of other neutralizing antibodies since they appear to compete with the neutralizing antibodies for the binding site. The possible role of such reactions in the formation of infectious immune complexes in EIA cannot be overlooked (96).
Fragmentation studies suggested that the antigenicity of gp90 was mainly localized on two fragments (12K and 20K) that reacted with reference EIA-positive horse serum and neutralizing or non-neutralizing MCAbs. All four antigenic sites of gp90 of prototype EIAV appeared to be present on these two fragments. Since the larger fragment (20K) was shown to be glycosylated, the carbohydrate moiety may play a role in masking potential epitopes on the protein backbone that are important for protective immunity.
Recent studies with EIAV suggest that both carbohydrate and protein moieties of gp90 contribute to its activity. The predominant reactivity, however, appear to be against its peptide epitopes (103).
Our studies of proteolytic fragmentation of the gp90 component of various EIAV isolates produced common 12K immunoreactive fragments.
This 12K proteolytic fragment(s) was reactive with non-neutralizing
MCAbs that reacted with 16 EIAV strains (54) and was also reactive with our reference EIA-positive horse serum. These results indicate the occurrence of conserved antigenic determinants on EIAV glycoproteins which may be beneficial as potential targets for developing a broadly effective immunogen against EIA. We have previously shown that epitopes on gp90 which are variable and highly antigenic may represent type-specific epitopes that react with neutralizing MCAbs (54). In the same study none of the isolates tested were neutralized by any of the four neutralizing MCAbs (54). In addition, each variant of EIAV studied in our laboratory was shown to be distinct by a variety of serological and biochemical procedures
(54, 102, 122, 129). Therefore, although, neutralizing MCAbs may provide a highly efficient mode of protection against viral infection
in general, they may not provide effective guidance in the search for a broadly effective vaccine against EIAV. Protection against a lethal virus infection by non-neutralizing MCAbs was shown with vesicular stomatitis virus (VSV) (82). Moreover, non-neutralizing
MCAbs that reacted with a cross-reactive epitope on the G-proteins of both VSV-Ind and VSV-NJ serotypes was able to confer protective immunity against challenge with either serotype (82). The fact that horses chronically infected with EIAV eventually develop an effective broad immunity against EIAV variants suggests the existence of common epitope(s) among EIAV variants. Such an approach might be advisable for developing a vaccine against EIAV. CHAPTER V
SUMMARY
Monoclonal antibodies were produced to the prototype cell-adapted
EIAV (Wyoming strain ), a lentivirus. Mice were immunized with different preparations of EIAV in order to generate MCAbs with different specifities. Twenty three hybridomas producing MCAbs were isolated and characterized. Ten were specific for the major envelope glycoprotein
(gp90), 8 for the other envelope glycoprotein (gp45) and 5 for the major group specific antigen (p26). Serologic reactivities were initially determined by ELISA. However, ultimate determination of MCAb reactivities with EIAV polypeptides was made by Western blot immunoassay and neutralization test. Monoclonal antibodies which reacted with p26 also reacted with another protein of an apparent molecular weight of
55,000. The studies suggested that this 55K protein is the gag precursor of EIAV, from which the internal proteins (p26, p!5, pll, and p9) are derived. All MCAbs directed against viral glycoproteins were highly specific, reacting only to gp90 or gp45. Additionally, no protein of molecular weight higher than gp90 could be detected by our panel of MCAbs.
To study the pattern, extent, and mechanism of antigenic variation
in EIAV, antigens of 16 EIAV isolates were analysed by MCAbs. Fifteen of these virus isolates were recovered from plasma samples obtained during discrete febrile episodes in experimental ponies. Thirteen of these isolates were generated in ponies infected in parallel with the same EIAV inoculum. Eight of these isolates, shown previously to be antigenically distinct in cross-neutralization tests, proved to be
128 129 distinct by analysis with MCAbs. On the basis of reactivity with the panel of MCAbs, the 16 different strains of EIAV were found to fall into
12 distinct antigenic groups. These data show that there is a large spectrum of possible EIAVserotypes and confirms that antigenic variation occurs with high frequency in EIAV. Moreover, the data also showed that variation is a rapid and random process, as no pattern of evolution of variants was evident by comparision of 13 isolates from parallel infections.
Heterogeneity of gp90 was documented by identification of nine different serologic groups. In spite of variations observed in epitopes
90-B, 90-C, 90-D, 90-E, and 90-F, a highly conserved epitope of gp90
(epitope 90-A) was recognized. In contrast, only two serologic groups of gp45 were identified. Therefore, conservation of EIAV antigenic determinants varied between the two glycoproteins; gp45 showed a moderate conservation, whereas gp90 had l i t t l econservation among the different EIAV isolates.
Our results documented epitope alterations in the virion surface glycoproteins among sequential virus isolates of EIAV during persistent infection. These were evident by subtle antigenic differences among isolates and by definitive antigenic alterations at one or more epitopes. Alterations observed in the glycoproteins involved both neutralizing and non-neutralizing epitopes during in vivo passage of the virus. These observed antigenic alterations of the surface glycoproteins confirmed previous reports which demonstrated genetic and structural changes of some of these isolates. Therefore, these studies indicate that a distinct (genetic, phenotypic, and antigenic) virus population may predomiMte during each febrile episode in a persistently 130 infected pony. Our studies showed that upon in vivo virus replication and during selection of one particular isolate from its progenitor, an epitope may lose its function in virus neutralization but retain its antigenic conformation. Our observations suggest that the production of non-neutralizing antibodies may play a role in the production of immune complex disease.
Our data support the hypothesis of antigenic variation as an important mechanism for the persistence and dissemination of EIAV.
Antigenic variation most likely involves frequent point mutations in the envelope gene which encodes the two main envelope glycoproteins, gp90 and gp45. When epitope alterations occur and are manifested by alterations in antigenicity of these glycoproteins, the emerging new variant may not be affected by immune pressures which had been effective against previous variants. The new variant then multiplies in an unrestricted fashion until immune responses to this novel variant effectively check virus replication. When this cycle repeats itself at frequent intervals, the result is the chronic (cyclic) form of EIA.
Although all 18 MCAbs bound to either gp90 or gp45 in Western blot assay only 4 of them (to gp90) neutralized infectivity. Monoclonal antibodies that reacted with conserved epitopes on gp90 and gp45 failed to neutralize EIAV. On the other hand, our studies suggest that the variable regions of the envelope gp90 were often neutralizing epitopes.
Each of the neutralizing antibodies was found to be directed against an epitope of gp90 as demonstrated by Western blot immunoassay. Therefore, the results showed that critical epitopes important for neutralization are clustered on gp90. Epitopes recognized by neutralizing antibodies were suggested to be continuous since they resisted treatment with sodium dodecyl sulfate. The 4 MCAbs that had neutralizing activity
reacted with gp90 in three different patterns suggesting different
epitope recognition. Therefore, it was possible to identify three
tentative antigenic sites on gp90 which function in neutralization.
However, data provided by additive ELISA, competitive ELISA, and Western
blot analysis indicated that antigenic reactivity of gp90 was localized
on at least four distinct epitopes, two of which were important in
neutralization. Our studies also revealed that these epitopes were
localized on a continuum of overlapping antigenic sites on gp90. On the
other hand, only two distinct non-overlapping epitopes were identified
on gp45. Competitive binding studies of neutralizing MCAbs and
reference EIA-positive horse serum delineated the presence of a
neutralization domain on gp90 that appears to be immunodominant both in naturally infected horses and in mice immunized with EIAV. Limited proteolytic fragmentation of the gp90 component of several serologically distinct EIAV isolates produced common 12K immunoreactive fragments that contained a conserved epitope (90-A). These results indicate the occurrence of conserved antigenic regions on EIAV glycoproteins as well as a neutralization domain on gp90, which can be used as potential targets for vaccine developments.
In summary our results documented epitope alterations in the two main surface glycoproteins (gp90 and gp45) of EIAV during persistent infection. Alterations observed involved both neutralizing and non-neutralizing epitopes. The variable regions of the envelope glycoproteins are often neutralizing epitopes. The specific changes noted for these glycoproteins resulted in altered antigenicity and signalled the emergence of antigenically distinct variants. These alterations are purportedly sufficient to allow the mutant viruses to temporarily escape immunological inactivation. Thus, antigenic variation is an important mechanism for persistence and dissemination of
EIAV. Most importantly however, the observed antigenic variation of
EIAV may represent a major challenge for developing a successful vaccine against EIA. Continuous epitopes which were targets for neutralization were demonstrated to cluster on a domain of gp90. In addition, a highly conserved region was also demonstrated on gp90 of different serologically distinct EIAV isolates. These studies suggest different approaches that could be followed to design a vaccine for EIA. For example, if conserved epitopes were identified that stimulate cross-protection, the solution would be straightforward. The importance of glycosylation of these epitopes needs to be clarified but previous studies showed that the predominant reactivity appeared to be against the peptide epitopes. Second, if a common determinant does not exist then a cocktail of the immunogenic fragments identified by neutralizing
MCAbs of different epidemiologically significant variants could be prepared and evaluated once such variants are identified. An elegant approach to design such a vaccine would involve the construction of
ISCOM (immunostimulating complex) containing neutralizing epitopes and other determinants localized on either the glycoproteins gp90, gp45 or the internal proteins such as p26 that stimulate cytotoxic T cells.
This novel immunogen would be capable of inducing both humoral and cellular immunity. If such an immunogen is capable of providing protection against infection and prevent latency then we would have generated an ideal vaccine for EIA which is the ultimate goal of immunization. CHAPTER VI
BIBLIOGRAPHY
1. Amborski, G.F., Jeffers, G., Amborski, R.L., and Issel, C.J. 1979. Equine infectious anemia virus: Development of a simple reproducible method for titrating infectivity of the cell-adapted strain. Am. J. Vet. Res. 40:302-304.
2. Archer, B.G., Crawford, T.B., McGuire, T.C., Frazier, M.E. 1977. RNA-dependent DNA polymerase associated with equine infectious anemia virus. J. Virol. 22:16-22.
3. Askonas, B.A., McMichael, A .J., and Webster, R.G. 1982. The immune response to influenza viruses and the problem of protection against infection, p. 159-188. A.S. Beare (ed.), Basic and applied influenza research. CRC Press, Inc., Boca Raton, Florida.
4. Bachrach, H.L., Moore, D.M., McKercher, P.D., and Polatnick, J. 1975. Immune and antibody responses to an isolated capsid protein of foot and mouth disease virus. J. Immunol. 115:1636-1641.
5. Banks, K.L., Henson, J.B ., and McGuire, T.C. 1972. Immunologically mediated glomerulitis of horses. I. Pathogenesis in persistent infection by equine infectious anemia virus. Lab. Invest. 26:701-707.
6. Benn, S., Rutledge, R., Folks, T., Gold, J ., Baker, L., McCormick, J., Feorino, P., Piot, P., Quinn, T., and Martin, M. 1985. Genomic heterogeneity of AIDS retroviral isolates from North America and Zaire. Science 230:949-954.
7. Blondel, B., Crainic, R., Fichot, 0 ., Bufraisse, G., Candrea, A., Diamond, D., Girard, M., and Horaud, F. 1986. Mutations conferring resistance to neutralization with monoclonal antibodies in type 1 poliovirus can be located outside or inside the antibody-binding site. J. Virol. 57:81-90.
8. Bolognesi, D., Collins, J., Leis, J., Moennig, V., Schafer, W., and Atkinson, P. 1975. Role of carbohydrate in determining the immunochemical properties of the major glycoprotein (gp71) of Friend murine leukemia virus. J. Virol. 16:1453-1463.
9. Bosselman, R.A., Van Straaten, F., Van Beveren, C., Verma, I.M. and Vogt, M. 1982. Analysis of the env gene of a molecularly cloned and biologically active Moloney Mink cell focus-forming proviral DNA. J. Virol. 44:19-31.
10. Breschkin, A.M., Ahern, J ., and White, D.0. 1981. Antigenic determinants of influenza virus hemagglutinin VIII. Topography of the antigenic regions of influenza virus haemagglutinin determined by competitive radioimmune assay with monoclonal
133 134
antibodies. Virology 113:130-140.
11. Bridgen, P.J., Cross, G.A.M., and Bridgen, J. 1976. N-terminal amino acid sequences of variant-specific antigens from Trypanosoma brucci. Nature (London) 263:613-614.
12. Brodeur, B.R., Tsang, P., and Larose, Y. 1984. Parameters affecting ascites tumor formation in mice and monoclonal antibody production. J. Immunol. Methods 71:265-272.
13. Burnette, W.N. 1981. "Western blotting" electrophoretic transfer of proteins from sodium dodecylsulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal. Biochem. 112:195-203.
14. Cao, Y., Schnurr, D.P. and Schmidt, N.J. 1984. Monoclonal antibodies for study of antigenic variation in coxsackievirus Type B4: Association of antigenic determinants with Myocarditic properties of the virus. J. Gen. Virol. 65:925-932.
15. Casey, J.M., Kim, Y., Anderson, P.R., Watson, K.F., Fox, J.L ., and Devare, S.G. 1985. Human T-Cell Lymphotropic virus type III. Immunologc characterization and primary structure analysis of the major internal protein, p 24. J. Virol. 55:417-423.
16. Charman, H.P., Bladen, S., Gilden, R.V., and Coggins, L. 1976. Equine infectious anemia virus: evidence favoring classification as a etrovirus. J. Virol. 19:1073-1079.
17. Cheevers, W.P., Ackley, C.M., and Crawford, T.B. 1978. Structural proteins of equine infectious anemia virus. J. Virol. 28:997-1001.
18. Cheevers, W.P., Archer, B.G., Crawford, T.B. 1977. Characterization of RNA from equine infectious anemia virus. J. Virol. 24:489-497.
19. Cheevers, W.P., and Mcguire, T.C. 1985. Equine infectious anemia virus: Immunopathogenesis and persistence. Rev. Infect. Dis. 7:83-88.
20. Cheevers. W.P., Robertson, G.A., Archer, B.G., Crawford, T.B. 1980. Orientation of structural polypeptides of equine infectious anemia virus. Intervirology 14:44-49.
21. Clegg, JCS. 1982. Glycoprotein detection in nitrocellulose transfers of electrophoretically separated protein mixtures using concanavalin A and peroxidase: application to arenavirus and flavivirus proteins. Anal. Biochem. 127:389-394.
22. Clements, J.E ., Pederson, F.S., Narayan, 0., and Haseltine, W.A. 1980. Genomic changes associated with antigenic variation of visna virus during persistent infection. Proc. Natl. Acad. Sci. 135 USA 77:4454-4458.
23. Clements, J.E ., D'Antonio, N., and Narayan, 0. 1982. Genomic changes associated with antigenic variation of visna virus. II. Common nucleotide sequence changes detected in variants from independent isolations. J. Mol. Biol. 158:415-434.
24. Cloyd. M.W., Bolognesi, D.P., and Bigner, D.D. 1977. Immnofluorescent analysis of expression of the RNA tumor virus major glycoprotein, gp 71, on surface of virus-producing murine and other mammalian species cell lines. Cancer Res. 37:922-930.
25. Cooper, N.R. 1979. Humoral immunity to viruses, p. 123-170. _l£: H. Fraenkel-Conrat and R.R. Wagner (ed.), Comprehensive Virology, Vol. 15, "Virus-Host Interactions, Immunity to viruses". Plenum, New York.
26. Coggins, L., and Norcross, N.L. 1970. Immunodiffusion reaction in equine infectious anemia. Cornell Vet. 60:330-335.
27. Coggins, L., Norcross, N.L. and Kemen, M. 1972a. The technique and applications of the immunodiffusion te st for equine infectious anemia. Proc. 3rd Intl. Conf. Equine Inf. Dis. Paris pp. 189-198.
28. Coggins, L., Norcross, N., Nusbaum, S. 1972b. Diagnosis of equine infectious anemia by immunodiffusion test. Am. J. Vet. Res. 33:11-18.
29. Crawford, T.B., Cheevers, W.P., Klevjer-Anderson, P., and McGuire,T.C. 1978. Equine infectious anemia: virion characteristics, virus-cell interaction and host responses, p. 727-749. Ijx : J. Stevens, G. Todaro, and F. Fox (eds.), Persistent viruses. ICN-UCLA Symposium on Molecular and Cellular Biology, Vol. II. Academic Press, Inc., New York.
30. Cross, G.A.M. 1975. Identification, purification and properties of clone-specific glycoprotein antigens constituting the surface coat of Trypanosoma brucei. Parasitology 71:393-417.
31. Daniels, C.A. 1975. Mechanisms of viral neutralization, p. 79-97. In: A.L. Notkins (ed.), Viral Immunology and ImmunopathoTogy. Academic Press, New York.
32. Dernick, R., Heukeshoven, J ., and Hilbrig, M. 1983. Induction of neutralizing antibodies by all three structural poliovirus polypeptides. Virology 130:243-246.
33. Dimmock, N.J. 1984. Mechanisms of neutralization of animal viruses. J. Gen. Virol. 65:1015-1022.
34. Donelson, J.E ., and Rice-Ficht, A.C. 1985. Molecular biology of Trypanosome antigenic variation. Microbiological review 49:107-125. 136
35. Douillard, J.Y., and Hoffman, T. 1983. Enzume-linked immunosorbent assay for screening monoclonal antibody production using enzyme-labeled second antibody. Meth. Enzymol. 92:168-174.
36. Emini, E.A., Kao, S.-Y., Lewis, A .J., Crainic, R. and Wimmer, E. 1983. Functional basis of poliovirus neutralization determined with monospecific neutralizing antibodies. J. Virol. 46:466-474.
37. Ey, P.L., Prowse, S .J., and Jenkin, C.R. 1978. Isolation of pure IgG., IgGg, and IgGp. immunoglobulins from mouse serum using protein A-Tfepharose. Tmmunochemistry 15:429-436.
38. Flamand, A., Wiktor, T .J., and Koprowski, H. 1980. Use of hybridoma monoclonal antibodies in the detection of antigenic differences between rabies and rabies-related virus proteins. II. The glycoprotein. J. Gel. Virol. 48:105-109.
39. Fleming, J.O ., Stohlman, S.A., Harmon, R.C., Lai, M.C., Frelinger, J.A., and Weiner, L.P. 1983. Antigenic relationships of murine coronaviruses: Analysis using monoclonal antibodies to JHM (MHV-4) virus. Virology 131:296-307.
40. Friguet, B., Djavadi-Ohaniance, L, Pages, J ., Bussard, A., and Goldberg, M. 1983. A convenient enzyme-linked immunosorbent assay for testing whether monoclonal antibodies recognize the same antigenic site. Application to hybridomas specific for the - Subunit of Escherichia coli Tryptophan Synthase. J. Immunol. Methods 60:351-358.
41. Fujimiya, Y., Perryman, L.E., and Crawford, T.B. 1979. Leukocyte cytotoxicity in a persistent virus infection: presence of direct cytotoxicity but absence of antibody-dependent cellular cytotoxicity in horses infected with equine infectious anemia virus. Infect. Immun. 24:628-636.
42. Galfre', G., and Milstein, C. 1981. Preparation of monoclonal antibodies: Strategies and procedures. Meth. Enzymol. 73:1-46.
43. Gallo, R.C. 1987. The AIDS virus. Scient. Am. 256:46-56.
44. Gerhard, W., Croce, C.M., Lopes, D., and Koprowski, H. 1978. Repertoire of antiviral antibodies expressed by somatic cell hybrids. Proc. Nat. Acad. Sci. USA 75:1510-1514.
45. Gimenez, H.B., Cash, P., and Melvin, W.T. 1984. Monoclonal antibodies to human respiratory syncytial virus and their use in comparison of different virus isolates. J. Gen. Virol. 65:963-971.
46. Gollins, S.W., and Porterfield, J.S. 1986. A new mechanism for the neutralization of enveloped viruses by antiviral antibody. Nature 321:244-246. 137
47. Gonda, M., Charman, H., Walker, J ., and Coggins, L. 1978. Scanning and transmission electron microscopic study of equine infectious anemia virus. Am. J. Vet. Res. 39:731-739.
48. Groth, S.F. and Scheidegger, D. 1980. Production of monoclonal antibodies: strategy and tactics. J. Immunol. Methods. 35:1-21.
49. Hahn, B.H., Shaw, G.M., Taylor, M.E., Redfield, R.R., Markham, P.D., Salahuddin, S.Z., Wong-Staal, F., Gallo, R.C., Parks, E.S., and Parks, W.P. 1986. Genetic variation in HTLV-11 I/LAV over time in patients with AIDS or at risk for AIDS. Science 232:1548-1553.
50. Hawkes, R. 1982. Identification of concanavalin A-binding proteins after sodium dodecyl sulfate-gel electrophoresis and protein blotting. Anal. Biochem. 123:143-146.
51. Henson, J.B ., Gorham, J.R ., Kobayashi, K., McGuire, T.C. 1969. Immunity in equine infectious anemia. J. Am. Vet. Med. Asso. 155(2):336-343.
52. Henson, J.B. and McGuire, T.C. 1974. Equine infectious anemia. Prog. Med. Virol. 18:143-159.
53. Hunsmann, G., Moenning, V., Pister, L., Seifert, E., and Schafer, W. 1974. Properties of mouse leukemia virus. VIII. The major viral glycoprotein of Friend leukemia virus. Seroimmunological, interfering and hemagglutinating capacities. Virology 62:307-318.
54. Hussain, K.A., Issel, C.J., Schnorr, K.L., Rwambo, P.M., and Montelaro, R.C. 1987. Antigenic analysis of equine infectious anemia virus (EIAV) variants by using monoclonal antibodies: epitopes of glycoprotein gp90 of EIAV stimulate neutralizing antibodies. J. Virol. 61:2956-2961.
55. Icenogle, J., Shiwen, H., Duke, G., Gilbert, S., Rueckert, R. and Anderegg, J. 1983. Neutralization of poliovirus by a monoclonal antibody: Kinetics and stoichiometry. Virology 127:412-425.
56. Ikeda, H., Pincus, T., Yoshiki, T., Strand, M., August, J.T., Boyse, E.A. and Mellors, R.C. 1974. Biological expression of antigenic determinants of murine leukemia virus proteins gp71 and p30. J. Virol. 14:1274-1280.
57. Ish ii, S., and Ishitani, R. 1975. Equine infectious anemia. Adv. Vet. Sci. Comp. Med. 19:195-222.
58. Ishitani, R. 1970. Equine infectious anemia. Natl. Inst. Anim. Health Q. 10:1-28.
59. Ishizaki, R., Green, R.W., and Bolognesi, D.P. 1978. The structural polypeptides of equine infectious anemia virus. Intervirology 9:286-294. 138
60. Issel, C.J., and Coggins, L. 1979. Equine infectious anemia: Current knowledge. J. Am. Vet. Med. Assoc. 174:727-733.
61. Issel, C .J., Montelaro, R.C., and Foil, L.D. 1986. Virology of equine retroviruses, p. 95-105. L.A. Salzman (ed.), Animal models of retrovirus infection and their relationship to AIDS. Academic Press, Orlando.
62. Kemeny, L.J., Mott, L.O., Pearson, J.E. 1971. Titration of equine infectious anemia virus: effect of dosage on incubation time and clinical signs. Cornell Vet. 61:687-695.
63. Kenel, S.J., Del Villano, B.C., Lewy, R.L., and Lerner, R.A. 1973. Properties of an oncornavirus glycoprotein: Evidence for its presence on the surface of virion and infected cells. Virology 55:464-475.
64. Kobayashi, K., and Kono, Y. 1967. Propagation and titratio n of equine infectious anemia virus in horse leukocyte culture. Natl. Inst. Anim. Health Q. 7:8-20.
65. Kohler, G., and M ilstein, C. 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature (London) 256:495-497.
66. Kohler, G., and Milstein, C. 1976. Derivation of specific antibody-producing tissue culture and tumor lines by cell fusion. Eur. J. Immunol. 6:511-519.
67. Kono, Y. 1969. Viremia and immunological response in horses infected with equine infectious anemia virus. Natl. Inst. An. Health Q. 9:1-9.
68. Kono, Y. 1973. Recurrence of equine infectious anemia. Approaches to an understanding of the mechanisms, p. 175-186. In proceedings of the 3rd International Conference on Equine Infectious Diseases, Karger, Basel.
69. Kono, Y., Fukunaga, Y., and Kobayashi, K. 1973a. Resistance of horses infected chronically with equine infectious anemia virus against reinfection. Natl. Inst. Anim. Health Q. 13:173-181.
70. Kono, Y., and Kobayashi, K. 1966. Complement-fixation test of equine infectious anemia. I. Specificity of the te st. Natl. Inst. Anim. Health Q. (Tokyo) 6:194-203.
71. Kono, Y., Kobayashi, K., and Fugunaga, Y. 1971. Serological comparison among various strains of equine infectious anemia virus. Arch. Gesamte Virusforsch. 34:202-208.
72. Kono, Y,, Kobayashi, K., and Fukunaga, Y. 1973b. Antigenic d rift of equine infectious anemia virus in chronically infected horses. Arch. Gesamte Virusforsch. 41:1-10. 139
73. Kono, S. and Yamamoto, H. 1970. Pathology of equine infectious anemia. Proposed classification of pathologic types of disease. Cornell Vet. 60:393-449.
74. Koprowski, H., Gerhard, W., and Croce, C.M. 1977. Production of antibodies against influenza virus by somatic cell hybrids between mouse myeloma and primed spleen cells. Proc. Nat. Acad. Sci. USA 74:2985-2988.
75. Koszinowiski, U.H., Allen H,, Gething, M.-J., Waterfield, M.D., and Klenk, H.-D. 1980. Recognition of viral glycoproteins by influenza A-specific cross-reactive cytotoxic T lyonphocytes. J. Exp. Med. 154:945-958.
76. Kreil, G. 1981. Transfer of proteins across membranes. Ann. Rev. Biochem. 50:317-348.
77. Kurrle, R., Wagner, H., Rollinghoff, and Rott, R. 1979. Influenza virusspecific T Cel 1-mediated cytotoxicity: Integration of the virus antigen into the target cell membrane is essential for target cell formation. Eur. J> Immunol. 9:107-111.
78. Kurz, C., Forss, S., Kupper, H.» Strohmaier, K., and Schaller, H. 1981. Nucleotide sequence and corresponding amino acid sequence of the genes for the major antigen of foot and mouth disease virus. Nucleic Acids Res. 9:1919-1931.
79. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.
80. Lamb, R.A., Zebedee, S.L. and Richardson, C.D. 1985. Influenza virus M? protein is an integral membrane protein expressed on the infectea-cell surface. Cell 40:627-633.
81. Laude, H., Chapsal, J.-M., Gelfi, J., Labiau, S., Grosclaude, J. 1986. Antigenic structure of transmissible gastroenteritis virus .1. Properties of monoclonal antibodies directed against virion proteins. J. Gen. Virol. 67:119-130.
82. Lefrancois, L. 1984. Protection against lethal viral infection by neutralizing and non-neutralizing monoclonal antibodies: distinct mechanisms of action in vivo. J. Virol. 51: 208-214.
83. Lernhardt, W., Anderson, J ., Coutinho, A., and Melchers, F. 1978. Cloning of murine transformed cell lines in suspension culture with efficiencies near 100 %. Expl. Cell Res. 111:309-316.
84. L ittlefield , J.W. 1964. Selection of hybrids from matings of fibroblasts in vitro and their presumed recombinants. Science 145:709-701.
85. Lowry, O.H. Rosebrough, N.J., Farr, A.L., and Randall, R.J. 1951. Protein measurement with the Folin phenol reagent. J. Biol. 140
Chem. 193: 265-275.
86. Lutley, R., Petursson, G., Palsson, P.A., Georgsson, G., Klein, J ., and Nathanson, N. 1983. Antigenic d rift in visna: virus variation during long term infection of kelandic sheep. J. Gen. Virol. 64:1433-1440.
87. Magrath, D .I., Evans, D.M., Perguson, M., Schild, G.C., Minor, P.D., Horaud, F., Crainic, R., Stenvik, M., and Hovi, T. 1986. Antigenic and molecular properties of type 3 poliovirus responsible for an outbreak of poliomyelitis in vaccinated population. J. Gen. Virol. 67:899-905.
88. Malmquist, W.A., Barnett, D., and Becvar, C.S. 1973. Production of equine infectious anemia antigen in persistently infected cell lines. Arch. Gesamte Virusforsch. 42:361-370.
89. Mandel, B. 1978. Neutralization of animal viruses, p. 205-268. In: M. Lauffer, K. Maramorosch, F. Bang and K. Smith (eds.), Advances in Virus Research, Vol. 23, Academic Press, New York.
90. Mandel, B. 1979. Interaction of viruses with neutralizing antibodies, p 37-121. _In: H. Fraenkel-Conrate and R.R. Wagner (eds.), Comprehensive Virology, Vol. 15, "Virus-Host Interactions, Immunity to Viruses", Plenum, New York.
91. Martins, J ., and Croce, C.M. 1978. Somatic cell hybrids producing antibodies specific for the tumor antigen of simian virus 40. Proc. Nat. Acad. Sci. USA 75:2320-2323.
92. Matheka, H.D., Coggins, L., Shively, J.N., Norcros, N. 1976. Purification and characterization of equine infectious anemia virus. Arch. Virol. 51:107-114.
93. McConnell, M., Katada, M., McConnell, S. 1977. Demonstration of equine infectious anemia virus in primary leukocytes culture by electron microscopy. Am. J. Vet. Res. 38:2067-2069.
94. McGuire, T.C., and Crawford, T.B. 1978. Induction of cell membrane antigen by equine infectious anemia virus. Am. J. Vet. Res. 39:385-386.
95. McGuire, T.C., Crawford, T.B., and Henson, J.B. 1971. Immunofluorescence localization of equine infectious anemia virus in tissues. Am. J. Pathol. 62:283-292.
96. McGuire, T.C., Crawford, T.B., and Henson, J.B. 1972. Equine infectious anemia. Detection of infectious virus-antibody complexes in the serum. Immunol. Comm. 1:545-551.
97. McGuire, T.C. and Henson, J.B. 1973. Equine infectious anemia-Pathogenesis of persistent viral infection. The Gustav Stern Symposium. Academic Press, New York. p. 229-247. mi
98. Montagnier, L., Dauguet, C., Axler, C., Chamaret, S., Gruest, J ., Nugeyre, M.T., Rey, F., Barre'-Sinoussi, F., and Chermann, J.C. 1984. A new type of retrovirus isolated from patients presenting with lymphodenopathy and acquired immune deficiency syndrome: structural and antigenic relatedness with equine infectious anemia virus. Annales Virologie (Institute Pasteur) 135E:119-134.
99. Montelaro, R.C., and Bolognesi, D.P. 1978. Structure and morphogenesis of type-C retroviruses. Adv. Cancer Res. 28:63-89.
100. Montelaro, R.C., Lohrey, N., Parekh, B., Blakeney, E.W. and Issel, C.J. 1982. Isolation and comparative biochemical properties of the major internal polypeptides of equine infectious anemia virus. J. Virol. 42:1029-1038.
101. Montelaro, R.C., West, M., and Issel, C.J. 1983. Isolation of equine infectious anemia virus glycoproteins. Lectin affinity chromatography procedures for high avidity glycoproteins. J. Virol. Methods 6:337-346.
102. Montelaro, R.C., Parekh, B., Orrego, A., and Issel, C.J. 1984a. Antigenic variation during persistent infection by equine infectiousanemia virus, a retrovirus. J. Bio. Chem. 259:10539-10544.
103. Montelaro, R.C. West, M., and Issel, C.J. 1984b. Antigenic reactivity of the major glycoprotein of equine infectious anemia virus, a retrovirus. Virology 136:368-374.
104. Montelaro, R.C., Issel, C.J., Payne, S., and Salinovich, 0. 1986. Antigenic variation during persistent infections by equine infectious anaemia virus, p. 41-56. J_n T.H. Birkbeck and C.W. Penn (eds.), Antigenic variation in infectious diseases. IRL Press, Washington, D.C.
105. Nakajima, H., Sugiura T., and Ushimi, C. 1978. Immunodiffusion test for the diagnosis of equine infectious anemia. J. Equine Med. and Surg. (Suppl.) 1:339-348.
106. Nakajima, H., Tajima, M., Tanaka, S., and Ushimi, C. 1969. Physicochemical studies of equine infectious anemia virus. III. Purification and electron microscopic observation of the virus. Arch. Gesamte Virusforsch. 28:348-360.
107. Nakajima, H., and Ushimi, C. 1971. Immunodiffusion studies of purified equine infectious anemia virus. Infet. Immun. 3:373-377.
108. Narayan, 0 ., Clements, J.E ., G riffin, D.E., and Wolinsky, J.S. 1981. Neutralizing antibody spectrum determines the antigenic profiles of emerging mutantsof visna virus. Infect. Immun. 32:1045-1050. m2
109. Narayan, 0., Griffin, D.E. and Chase, J. 1977a. Antigenic shift of visna virus in persistently infected sheep. Science 197:376-378.
110. Narayan, 0., G riffin, D.E., and Clements, J.E. 1978. Virus mutation during 'slow infection': Temporal development and characterization of mutants of visna virus recovered from sheep. J. Gen. Virol. 41:343-352.
111. Narayan, 0 ., G riffin, D.E. and Silverstein, A.M. 1977b. Slow virus infection: replication and mechanisms of persistence of visna virus in sheep. J. Infect. Dis. 135:800-806.
112. Nishimura, M., and Nakajima, H. 1984. Structural proteins of equine infectious anemia virus and their antigenic activity. Am. J. Vet. Res. 45:5-10.
113. Nowinski, R.C., Lostrom, M. E ., Tam, M.R., Stone, M.R., and Burnette, W.N. 1979. The isolation of hybrid cell lines producing monoclonal antibodies against the pl5(E) protein of ecotropic murine leukemia viruses. Virology 93:111-126.
114. Nunberg, J.H., Rodgers, G., Giobert, J.H., and Snead, R.M. 1984. Method to map antigenic determinants recognized by monoclonal antibodies: localization of a determinant of virus neturalization on the feline leukemia virus envelope protein gp70. Proc. Natl. Acad. Sci. 81:3675-3679.
115. Oi, V.T., and Herzenberg, L.A. 1980. Immunoglobulin producing hybrid cell lines, p 351-372. Jin B.B. Mishell and S.M Shiigi (ed.), Selected Methods in Cellular Immunology. Freeman, San Francisco.
116. Orrego, A., Issel, C.J., Montelaro, R.C., and Adams, W.V. 1982. Virulence and in vitro growth of a cell adapted strain of equine infectious anemia virus after serial passage in ponies. Am. J. Vet. Res. 43:1556-1560.
117. Orrego, A. 1983. Equine infectious anemia virus: I. Generation of virus variant candidates. II. Attempts to titr a te by three different methods. Ph.D. dissertation, Louisiana State University, Baton Rouge, LA.
118. O'Sullivan, M.J., Gnemmi, E., Morris, D., Chieregatti, G., Simmonds, A.D., Simmons, M., Bridges, J.W. and Marks, V. 1979. Comparison of two methods of preparing enzyme-anti body conjugates: application of these conjugates for enzyme immunoassays. Anal. Biochem. 100:100-108.
119. Palese, P., and Young, F. 1982. Variation of influenza A, B, and C viruses. Science. 215:1468-1472.
120. Parekh, B., Issel, C.J., and Montelaro, R.C. 1980. Equine infectious anemia virus, a putative lentivirus, contains 1A3
polypeptides analogous to prototype-C oncornaviruses. Virology 107:520-525.
121. Payne, S., Parekh, B., Montelaro, R.C., and Issel, C.J. 1984. Genomic alterations associated with persistent infections by equine infectious anemia virus, a retrovirus. J. Gen. Virol. 65:1395-1399.
122. Payne, S.L., Salinovich, 0 ., Nauman, S.M., Issel, C.J., and Montelaro, R.C. 1987. Course and extent of variation of equine infectious anemia virus during parallel persistent infections. J. Virol. 61:1266-1270.
123. Pearson, J ., Coggins, L., and House, C. 1975. Protocol for the immunodiffusion (Coggins) te st for equine infectious anemia. Am. Assoc. Vet. Lab. Diag. Proc. 18:1-13.
124. Portetelle, D., Bruck, C., Mammerickx, M., and Burney, A. 1980. In animals infected by bovine leukemia virus (BLV) antibodies to envelope glycoprotein gp51 are directed against the carbohydrate moiety. Virology 105:223-233.
125. Ratner, L., Gallo, R.C., and Wong-Staal, F. 1985. HTLV-III, LAV, ARV are variants of same AIDS virus. Nature 313:636-637.
126. Rice, N.R., Simek, S., Ryder, O.A., and Coggins, L. 1978. Detection of proviral DNA in horse cells infected with equine infectious anemia virus. J. Virol. 26:577-583.
127. Rifkin, D.B. and Compans, R.W. 1971. Identification of spike proteins of Rous sarcoma virus. Virology 46:485-489.
128. Roth, E., Coggins, L., Patten, V. 1971. Evaluation of the immunodiffusion test for the diagnosis of equine infectious anemia. U.S. Animal Health Assoc. Proc. 75:222-231.
129. Salinovich, 0 ., Payne, S.L., Montelaro, R.C., Hussain, K.A., Issel, C.J., and Schnorr, K.L. 1986. Rapid emergence of novel antigenic and genetic variants of equine infectious anemia virus during persistent infection. J. Virol. 57:71-80.
130. Schafer, W., and Bolognesi, D.P. 1977. Mammalian C-type oncornaviruses: Relationships between viral structure and cell-surface antigens and their possible significance in immunological defense mechanisms. Contemp. Topics Immunobiol. 6:127-167.
131. Schmerr, M., M iller, J ., and Van Der Maaten, M. 1981. Antigenic reactivity of a soluble glycoprotein associated with bovine leukemia virus. Virology 109:431-434.
132. Scott, J.V., Stowring, L., Haase, A.T., Narayan, 0., and Vigne, R. 1979. Antigenic variation in visna virus. Cell 18:321-327. 144
133. Sentsui, H., and Kono, Y. 1976. Hemagglutination by equine infectious anemia virus. Infect. Immun. 14:325-331.
134. Shulman, M., Wilde, C.D., and Kohler, G. 1978. A better cell line for making hybridomas secreting specific antibodies. Nature 276:269-270.
135. Steeves, R.A., Strand, M., and August, J.T. 1974. Structural proteins of mammalian oncogenic RNA viruses: Murine leukemia virus neutralization by antisera prepared against purified envelope glycoproteins. J. Virol. 14:187-189.
136. Stein, C.D., and Gates, D.W. 1950. The neutralization effect of antiserum from recovered carriers of equine infectious anemia on the virus of the disease. Vet. Med. 45:152-156.
137. Stephens, R.M., Casey, J.W., Rice, N.R. 1986. Equine infectious anemia virus gag and pol genes: Relatedness to visna and AIDS virus. Science 231:589-594.
138. Stowring, L., Haase, A.T., and Charman. H.P. 1979. Serological definition of the lentivirus group of retroviruses. J. Virol. 29:523-528.
139. Tajima, M., Nakajima, H., Ito, Y. 1969. Electron microscopy of equine infectious anemia virus. J. Virol. 4:521-527.
140. Tanaka, K., and Sakaki, K. 1962. Neutralization test on serum from horses infected with the virus of equine infectious anemia. Natl. Inst. Anim. Health Q. 2:128-139.
141. Thormar, H . , Barshatzky, M.R., and Kozlowski, P.B. 1983. The emergence of antigenic variation is a rare event in long term visna virus infection in vivo. J. Gen. Virol. 64:1427-1432.
142. Thorpe, R., Bird, C.R., and Spitz, M. 1984. Immunoblotting with monoclonal antibodies: Loss of immunoreactivity with human immunoglobulins arises from polypeptide chain separation. J. Immunol. Methods 73:259-265.
143. Towbin, H., and Gordon, J. 1984. Immunoblotting and dot immunobinding Current status and outlook. J. Immunol. Methods 72:313-340.
144. Underwood, P.A. 1984. An antigenic map of the haemagglutinin of the influenza Hong Kong subtype (H3N2) constructed using mouse monoclonal antibodies. Mol. Immunol. 21:663-671.
145. Van Eldik, L., Paulson, J ., Green, R., and Smith, R. 1978. The influence of carbohydrate on the antigenicity of the envelope glycoprotein of avian myeloblastosis virus and B77 avian sarcoma virus. Virology 86:193-204. 145
146. Voller, A., Bidwell, D., and B artlett, A. 1980. Enzyme-linked immunosorbent assay, p. 359-371. In N. R. Rose and H. Friedman (ed.), Manual of Clinical ImmunFTogy. American Society for Microbiology, Washington, D.C.
147. Watson, R.J., Weis, J.H., Salstrom, J.S ., and Enquist, L.W. 1982. Herpes simplex virus type-1 glycoprotein D gene: nucleotide sequence and expression in Escherichia Coli. Science 218:381-384.
148. Webster, R.G., and Laver, W.G.. 1980. Determination of the number of non-overlapping antigenic areas on Hong Kong (H3N2) influenze virus HA with monoclonal antibodies and the selection of variants with potential epidemiological significance. Virology 104:139-148.
149. Webster, R.G., Laver, W.G., Air, G.M., Shild, G.C. 1982. Molecular mechanisms of variation in influenza viruses, nature 296:115-121.
150. Weiland, F., Matheka, H., Coggins, L., and Hartner, D. 1977. Electronmicroscopic studies on equine infectious anemia virus (EIAV). Arch. Virol. 55:335-340.
151. Weiss, R. 1984. Pathogenesis of retrovirus-induced disease, p. 785-998. 2lL R* Weiss, N. Teich, H. Varmus, and J. Coffin (eds.), Molecular biology of tumor viruses "RNA tumor viruses." Cold Spring Harbor Laboratory, New York.
152. Wong-Staal, F., Shaw, G.M., Hahn, B.H., Salahuddin, S.Z., Popovic, M., Markham, P., Redfield, R., and Gallo, R.C. 1985. Genomic diversity of human T-lymphotropic virus type III (HTLV-III). Science 229:759-762.
153. Yaniv, A., Dahlberge, J ., Gazit, A., Sherman, L., Chiu, I-M., Tronick, S.R., and Aaronson, S.A. 1986. Molecular cloning and physical characterization of integrated equine infectious anemia virus: molecular and immunologic evidence of its close relationship to ovine and caprine lentiviruses. Virology 154:1-8.
154. Yewdell, J.W. and Gerhard, W. 1981. Antigenic characterization of viruses by monoclonal antibodies. Ann. Rev. Microbiol. 35:185-206. CHAPTER VII APPENDIX
LIST OF ABBREVIATIONS
ABTS 2,2'-azino-bis(3-ethylbenz-thiazoline-6-sulfonic acid),
diammonium salt
AGID Agar gel immunodiffusion
AI Additivity index
AIDS Acquired immune deficiency syndrome
BSA Bovine serum albumin
CAEV Caprine a rth ritis encephalitis virus
DMSO Dimethyl sulfoxide
EIA Equine infectious anemia
EIAV Equine infectious anemia virus
ELISA Enzyme-linked immunosorbent assay
FCA Freund'S complete adjuvant
FCS Fecal calf serum
FDD Fetal donkey dermal
FEK Fetal equine kidney
FeLV Feline leukemia virus
HAT Hypoxanthi ne/Ami noteri n/Thymi di ne
HEPES N-2-hydroxyethylpiperizine-N'-2-ethane sulfonic acid
HGPRT Hypoxanthine-guanine phosphoribosyl transferase
HIV Human immunodeficiency virus
HT Hypoxanthi ne/Thymi di ne
HTLV-I Human T-cell leukemia virus type I
HTLV-II Human T-cell leukemia virus type II
HTLV-111 Human T-cell lymphotropic virus type III
IP Intraperitoneally 146 IV Intravenously
K Kilodalton
MCAb Monclonal antibody
MEM Minimal essential medium
MLV Murine leukemia virus
NP-40 Nonidet P-40
ONPG O-nitrophenyl-B-D-galactoside
PAGE Polyacrylamide gel electrophoresis
PEG Polyethylene glycol
RPMI 1640 Rosewell Park Memorial Institute 1640
SC Subcutaneously
SDS Sodium dodecyl sulfate
TEMED N,N,N'N'-tetramethylene diamine
TK Thymidine kinase
TRIS Tris (hydroxymethyl) ami noethane CHAPTER VIII VITAE NAME: KHALID ABDULLAH HUSSAIN
DATE AND PLACE OF BIRTH: July 1, 1953 Mosul, IRAQ
CITIZENSHIP: IRAQ
HOME ADDRESS: Hai 14 Tammouz HOME PHONE: (011-964-1) 415-6964 House No. 2221/5/3 Baghdad, IRAQ
MARITAL STATUS: Married CHILDREN: Saad, 1983
COLLEGE AND UNIVERSITY EDUCATION:
Institution Dates Speciality Degree College of Vet. Med. 1973-1977 Veterinary B.V.M.& S. Univer. of Baghdad Medicine Baghdad, IRAQ
School of Vet Medicine 1980-1983 Virology M.S. Dept, of Vet. Micro. & Pathology Washington State Univer. Pullman, WA
School of Vet. Medicine 1983-1988 Virology Ph.D. Dept, of Vet. Micro. & Para. Louisiana State Univ. Baton Rouge, LA
WORK EXPERIENCE
Institution Dates Speciality Title
Military Service 1977-1978 General Veterinarian Vet. Med.
Col lege of Vet. Med. 1979-1980 Virology Teaching Dept, of Vet. Micro/ Assistant Virology Section Univer. of Baghdad Baghdad, IRAQ
PUBLICATIONS:
1. Hussain, K.A. C.J. Issel, K.L. Schnorr, P.M. Rwambo, and R.C. Montelaro. 1987. Antigenic analysis of equine infectious anemia virus (EIAV) variants by using monoclonal antibodies: epitopes of glycoprotein gp90 of EIAV stimulate neutralizing antibodies. J. Virol. 61:2956-2961.
148 149
2. Hussain, K.A., C.J. Issel, K.L. Schnorr, P.M. Rwambo, M. West, and R.C. Montelaro. 1988. Antigenic mapping of the envelope proteins of equine infectious anemia virus: identification of a neutralization domain and a conserved region on glycoprotein 90. Arch. Virol. (In press).
3. Salinovich, 0 ., S.L. Payne, R.C. Montelaro, K.A. Hussain, C.J. Issel and K.L. Schnorr. 1986. Rapid emergence of novel antigenic and genetic variants of equine infectious anemia virus during persistent infection. J. Virol. 57: 71-80.
ABSTRACTS:
1. Hussain, K.A., C.J. Issel, R.C. Montelaro, and K.L. Schnorr. 1984. Production and Characterization of Monoclonal Antibodies Against Equine Infectious Anemia Virus. Conference of Research Workers in Animal Diseases, Chicago Illin o is, Abstract number 63.
2. 0. Salinovich, K. Hussain, R. Montelaro, C. Issel, and K. Schnorr. 1985. Antigenic and structural variation of the envelope glycoprotein of equine infectious anemia virus during persistent infection. American society for Virology, University of New Mexico, Albuquerque, New Mexico., Workshop #32, paper number 7.
3. Hussain, K.A., C.J. Issel, R.C. Montelaro, and K.L. Schnorr. 1986. Epitopes of glycoprotein 90 (gp90) of equine infectious anemia virus (EIAV) stimulate neutralizing antibodies: Antigenic analysis. The annual meeting of the Animal Disease Research Workers in the Southern States (ADRWSS) at the university of Florida, Gainesville, Florida, Abstract number 11.
4. Hussain, K.A., C.J. Issel, R.C. Montelaro, and K.L. Schnorr. 1986. Antigenic analysis of equine infectious anemia virus using monoclonal antibodies: Epitopes of gp90 stimule neutralizing antibodies. American Society for Virology, University of California, Santa Barbara, California. Workshop #10, paper number.
5. Hussain, K.A., C.J. Issel, R.C. Montelaro and K.L. Schnorr. 1986. Topographical analysis of equine infectious anemia virus (EIAV) glycoproteins. American Society for Microbiology South Central Branch, Shreveport, Louisiana. Abstract number A3-5.
6. Arnizaut, A.B., B. Betterbed, C.J. Issel, K.A. Hussain and K.L. Schnorr. 1986. Immunofluorescence assay using polyclonal and monoclonal antibodies to detect cells infected with equine infectious anemia virus (EIAV). American Society for Microbiology South Central Branch, Shreveport, Louisiana. Abstract number B3-2.
7. Rwambo, P.M., C.J. Issel, R.C. Montelaro, and K.A. Hussain. 1986. Distinct serotypes of the equine lentivirus isolated from a pony with chronic equine infectious anemia (EIA). American Society for Microbiology South Central Branch, Shreveport, Louisiana. Abstract number B3-1. 150
8. Rwambo, P.M., C.J. Issel, K.A. Hussain, and R.C. Montelaro. 1987. In vitro isolation of a variant of equine infectious anemia virus (EIAV) using neutralizing serum. The 68th Annual meeting of the Conference of Research Workers in Animal Disease, Chicago, Illinois, Abstract number 346. DOCTORAL EXAMINATION AND DISSERTATION REPORT
Candidate: k H/\LID ABDULLAH HUSSAIN
Major Field: VETERINARY MEDICAL SCIENCES (VIROLOGY)
Title of Dissertation: ANTIGENIC ANALYSIS OF EQUINE INFECTIOUS ANEMIA VIRUS
USING MONOCLONAL ANTIBODIES
Approved:
M ajor Professor and Chairman
SychoolDean of the Graduate:e SychoolDean
EXAMINING COMMITTEE: J) azct
Date of Examination:
3/8/1988