Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations
1996 Interactions between bovine retroviruses and the host immune system Jeffrey Allen Isaacson Iowa State University
Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Animal Sciences Commons, Immunology and Infectious Disease Commons, Medical Immunology Commons, and the Veterinary Medicine Commons
Recommended Citation Isaacson, Jeffrey Allen, "Interactions between bovine retroviruses and the host immune system " (1996). Retrospective Theses and Dissertations. 11534. https://lib.dr.iastate.edu/rtd/11534
This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
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 copyright material had to be removed, a note wUl indicate the deletion.
Oversize materials (e.g., maps, drawmgs, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 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. UMI A Bell & Howell Information Company 300 North Zed) Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600
Interactions between bovine retroviruses and the host immune system
by
Jeffrey Alien Isaacson
A dissertation submitted to the graduate faculty in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY
Major; Immunobiology Major Professor. James A. Roth
Iowa State University Ames, Iowa 1996 XJMI NiJinber: 9712562
UMI Microform 9712562 Copyright 1997, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 ii
Graduate College Iowa State University
This is to certify that the Doctoral dissertation of Jeffrey Allen Isaacson has met the dissertation requirements of Iowa State University
Signature was redacted for privacy. Major Professor Signature was redacted for privacy.
Signature was redacted for privacy.
For tl mduate College iii
TABLE OF CONTENTS
LIST OF TABLES vii
LIST OF FIGURES vlii
ABSTRACT ix
CHAPTER 1. GENERAL INTRODUCTION 1
Dissertation Organization 1 Research Sunnnnary 1
CHAPTER 2. LITERATURE REVIEW: IMMUNOPATHOGENESIS OF BOVINE IMMUNODEFICIENCY VIRUS AND BOVINE LEUKEMIA VIRUS INFECTIONS IN CATTLE 3
Introduction 3 General nature of retroviruses 3 Retroviruses and the host immune system 5 Discovery of bovine retroviruses 6 Bovine Immunodeficiency Virus (BIV) 7 BIV genetics and replication 7 Host immune responses to BIV infection 9 Prevalence of BIV infection 10 Studies of BIV pathogenesis 11 Effects of BIV infection on the host immune system 14 Summary 17 Bovine Leukemia Virus (BLV) 17 General virology 17 Clinical manifestations of BLV infection 20 Host immune responses to BLV infection 22 Detection and prevalence of BLV infection 24 iv
Pathogenic mechanisms In BLV infection 26 Effects of BLV infection on the host immune system 34 Summary 44 Conclusions 44 References 45
CHAPTER 3. LOSS OF GAG-SPECIFIC ANTIBODIES IN CATTLE EXPERIMENTALLY INFECTED WITH BOVINE IMMUNODEFICIENCY-LIKE VIRUS 61
Abstract 61 Introduction 62 Materials and Methods 64 Animal inoculation, serum collection, and clinical monitoring 64 Virus isolation 65 BIV antigens and immunologic reagents 66 Westem immunoblotting 67 Immunoprecipitation of Gag3 67 Results 68 Virus recovery and clinical observations 68 Pattems of antibody reactivity to BIV Gag3 recombinant protein by immunoblotting 70 Antibody reactivity to Env8 72 Immunoprecipitation of GagS 72 Discussion 76 Acknowledgements 79 References 79
CHAPTER 4. EFFECTS OF LONG-TERM INFECTION WITH BOVINE IMMUNODEFICIENCY VIRUS AND/OR BOVINE LEUKEMIA VIRUS ON HUMORAL AND LYMPHOCYTE PROLIFERATIVE RESPONSES IN CATTLE 84
Abstract 84 Abbreviations 85 V
introduction 85 Materials and Methods 88 Animals and virus inoculations 88 Peripheral blood cell counts 89 Vaccinations 90 Serology 91 Isolation of PBMC 92 Lymphocyte proliferation assays 93 Mixed leukocyte reactions 93 Statistical analysis 94 Results 95 Confimnation of viral infection 95 Clinical monitoring 96 Peripheral blood counts 96 Humoral responses to tetanus toxoid 98 Lymphocyte proliferation 100 Effect of BLV infection on immune responses to BHV1 104 Discussion 104 Acknowledgements 110 References 110
CHAPTER 5. INCREASED EXPRESSION OF MHC CLASS II AND CD25 MOLECULES ON LYMPHOCYTES FROM NON-LYMPHOCYTOTIC, BLV- INFECTED CATTLE 114
Abstract 114 Abbreviations 115 Introduction 115 Materials and Methods 118 Animal Inoculations 118 Peripheral blood cell counts 118 Isolation of PBMC 118 vi
Culture and han/esting of cells 119 Monoclonal antibodies (Mabs) 120 Cell staining and flow cytometry 120 Statistical analyses 121 Results 121 Confirmation of BLV infection 121 Comparison of leukocyte populations in freshly isolated PBMC from BLV-lnfected vs. control animals 122 Expression of activation markers MHCII and CD25 on lymphocyte subsets in freshly isolated PBMC 124 Expression of MHCII by lymphocytes In cultured PBMC 124 Expression of CD25 by lymphocytes in cultured PBMC 126 Discussion 128 Acknowledgements 134 References 134
CHAPTER 6. SUMMARY AND GENERAL CONCLUSIONS 138
Summary of Experimental Results 138 Practical Implications 142 Theoretical Considerations 144 Recommendations for Future Research 148 Mechanism(s) of BIV Gag-specific antibody loss 148 Effects of BIV on acquired immune responses 149 Mechanism(s) of BLV-induced enhancement of antibody responses 150 References 152 VII
LIST OF TABLES
CHAPTER 2
Table 1. Prevalence of BIV seroposltlvrty in the United States and Canada 12
Table 2. Reported B cell percentages and/or absolute numbers in control vs. BLV-infected cattle with or without PL 35
Table 3. Reported T cell percentages and/or absolute numbers in control vs. BLV-infected cattle with or without PL 39
CHAPTER 3
Table 1. Recovery of virus from PBMC of cattle experimentally infected with BIV or with BIV and BLV 69
CHAPTER 4
Table 1. Comparison of cell numbers in peripheral blood for the fifth year p.i. 97
CHAPTER 5
Table 1. Effects of BLV infection on numbers of white blood cells and lymphocyte subsets in peripheral blood 123 VIII
LIST OF FIGURES
CHAPTER 3
Rgure 1. Seroreactivity to the 65 kDa GagS recombinant fusion protein in cattle experimentally infected with BIV 71
Figure 2. Persistence of antibody reactivity to a BIV env-encoded protein 74
Figure 3. Immunopreclpitation of IgG immune complexes from BIV- infected cattle 75
CHAPTER 4
Figure 1. Effect of experimental BIV and/or BLV infection on humoral responses to tetanus toxoid (TT). 99
Figure 2. Effect of bovine retroviral infections on lymphocyte proliferation in vitro 102
Figure 3. Effect of BLV infection on humoral and lymphocyte proliferative responses to BHV-1 vaccination 105
CHAPTER 5
Figure 1. Expression of MHCII and CD25 on lymphocyte subsets in freshly isolated PBMC from five BLV-infected cattle and three uninfected controls 125
Figure 2. Effect of BLV infection on MHCII expression by cultured lymphocytes 127
Figure 3. Effect of BLV infection on CD25 expression by cultured lymphocytes 129 ix
ABSTRACT
Immunological aspects of experimental infection of cattle with bovine Immunodeficiency virus (BIV) and/or bovine leukemia virus (BLV) were studied. The specificity and kinetics of antibody responses to BIV antigens were detemnined by immunoblot analysis using two BIV recombinant fusion proteins expressed in E. coli. Sera from all eight BlV-lnfected cattle reacted to a p26-containing protein by four weeks post-inoculation (p.L). but this reactivity decreased dramatically in seven cattle by six months post-inoculation. In contrast, reactivity to a protein containing the amino terminus of BIV gp42 persisted in all animals throughout the four year study. BIV could be isolated from peripheral blood mononuclear cells (PBMC) of all animals during the period of declining p26-specific antibodies. However, there was no obvious increase In virus recovery during this period, nor were any p26-containing circulating immune complexes detected by immunoprecipitatlon. Therefore, the antibody loss did not appear to be attributable to an increasing p26 antigenemia. These findings also suggest that seroepidemiological surveys that rely on detection of antibodies to p26 may underestimate the true prevalence of BIV. Because there was no evidence of immunosuppressive disease In these cattle, it was hypothesized that a suppression of antigen-specific immune responses contributed to the loss of p26-speclfic antibodies. Humoral responses to vaccination with tetanus toxoid (TT) were detemnined, as were lymphocyte proliferative responses to TT, allogeneic cells, and concanavalln A, Cattle infected with BIV and BLV, or BLV only, were also Included in this study to examine possible viral Interactions, The only consistent BIV effect was a reduction in responses to concanavalin A. However, BLV-infected cattle had significantly enhanced secondary antibody responses to TT and bovine herpes vims-l. Although only one BLV-infected animal was persistently X lymphocytotic, circulating B cell and CD4+ T cell numbers were generally increased in BLV-infected cattle. There was also evidence for increased lymphocyte activation in BLV-infected cattle, including: 1) increased percentage of B cells expressing MHCII in freshly isolated PBMC; 2) increased percentage of B cells expressing CD25 and; 3) increased percentage of T cells expressing MHCII following short term culture of PBMC. These findings suggest that BLV may induce a generalized lymphocyte activation. 1
CHAPTER 1. GENERAL INTRODUCTION
Dissertation Organization
This dissertation Is composed of a general introduction, a literature review on the Immunopathogenesis of bovine Immunodeficiency virus (BIV) and bovine leukemia virus (BLV) Infections, three manuscripts, and a section summarizing experimental results and their implications. The manuscript entitled "Loss of gag- specific antibody reactivity In cattle experimentally infected with bovine Immunodeficiency-like virus" was published in Viral Immunology (Volume 8, pages 27-36, 1995, used by permission). The other two manuscripts, as well as part of the literature review, will be submitted for publication. The references cited in the literature review and the three papers appear at the ends of respective chapters.
Research Summary
All of the studies described In this dissertation were conducted using a group of cattle experimentally Infected with BIV only (n = 3), BIV and BLV (n = 5), BLV only (n = 5), or sham-Inoculated controls (n = 3). Initially, BIV recombinant proteins were expressed and used to characterize the specificity and kinetics of antibody responses to BIV. After finding that Gag-speclfic antibodies declined dramatically in most Infected cattle by six months post-inoculation, it was hypothesized that this may have been a reflection of early suppression of acquired Immune reponses. It was also of interest to determine whether there was any interaction between BIV and BLV in temris of effects on host immune responses. Thus, the research focus shifted to the study of the effect of BIV and/or BLV infection 2 on host immune responses to non-retroviral antigens. There were no dramatic BIV effects, but it was serendipitousiy discovered that cattle infected with BLV had significantly enhanced antibody responses to two recall antigens. Although both B cell and T cell numbers were also Increased In these cattle, it was hypothesized that lymphocyte functions may also have been altered by BLV infection. This hypothesis was verified by showing that an increased percentage of B cells expressed MHCII molecules in freshly isolated peripheral blood mononuclear cells (PBMC) from BLV infected cattle. Moreover, in cultured PBMC, it was shown that an Increased percentage of T cells from BLV-infected animals expressed MHCII molecules, and more B cells also expressed the IL-2 receptor alpha chain. Thus, while BIV appeared to have relatively mild effects on bovine immune responses, BLV affected both numbers and functions of lymphocytes in infected cattle. Practical and theoretical implications of these findings are discussed in the conclusion of this dissertation. 3
CHAPTER 2. LITERATURE REVIEW: IMMUNOPATHOGENESIS OF BOVINE IMMUNODEFICIENCY VIRUS AND BOVINE LEUKEMIA VIRUS INFECTIONS IN CATTLE
Introduction
General nature of retroviruses
Retroviruses are enveloped RNA viruses that are distinguished by their use of an RNA-dependent DNA polymerase, reverse transcriptase, which generates a
DNA proviral form which integrates into the host genome. Both genomic viral RNA and viral mRNA coding for viral proteins can then be transcribed from the provirus by the host RNA polymerase. Endogenous retroviruses are a part of the genome of many mammalian species, while exogenous retroviruses, which spread horizontally between individual hosts, are important pathogens of both humans and domestic animals. Retroviruses have also been of great use in both biotechnology and molecular biology. Retroviral reverse transcriptase is commonly used to make cDNA copies of cellular RNA, and modified retroviruses have much potential as agents to transfer genes in the therapy of genetic diseases.
Retroviruses have been historically classified into three subfamilies; the
Lentivirinae, Oncovirinae, and Spumavirinae, In 1991, retroviruses were re classified into seven separate genera according to the Fifth Report of the
International Committee on Taxonomy of Viruses (Coffin, 1992). The lentiviruses and spumaviruses remained as separate genera, but the Oncovirinae were divided into five genera according to differences in morphology, genetics, and pathogenic mechanisms. Retroviruses have also been grouped according to the complexity of their life cycle (Cullen, 1992). Murine and avian leukemia viruses are considered 4
"simple" retroviruses, while the lentiviruses, spumaviruses, and the bovine leukemia virus (BLV)/human T cell leukemia virus (HTLV) group are considered to be "complex". The complex viruses are characterized by their possession of genes coding for two nuclear regulatory proteins, one which transactivates viral gene expression and another which promotes the expression of viral structural mRNA species, thus dividing the replication cycle into early and late phases. In contrast, simple retroviruses may require only gag, pol, and env genes for their replication.
The lentiviruses comprise the human, simian, feline, and bovine immunodeficiency viruses (HIV, SIV, FIV, and BIV), as well as equine infectious anemia virus (EIAV), maedi-visna virus, and caprine arthritis encephalitis virus.
These "slow" viral infections are usually characterized by long incubation periods with gradual development of disease, although some ElAV-infected horses have an acute course (Carpenter and Alexanderson, 1992). Another exception is
Jembrana disease virus, which can cause an acute and often fatal disease in Bali cattle, and has recently been identified as a lentivirus (Chadwick et al., 1995). Most genera of cancer-causing retroviruses either preferentially insert into the host genome where they can activate host proto-oncogenes, or themselves possess oncogenes which code for growth factor-like proteins that promote tumor formation.
In contrast, the HTLV/BLV viruses do not appear to have preferred sites of integration, and lack classical viral oncogenes. Although the mechanism(s) are incompletely understood, it is thought that BLV and HTLV contribute to tumor fomnation by producing regulatory proteins which not only increase viral transcription, but also activate host cell proliferation, leading to tumor formation in a relatively small number of hosts. 5
Retroviruses and the host immune system
Because they generally infect cells important in host defense, the life cycle of exogenous retroviruses is inextricably linked to the immune responses of the host.
Narayan and Clements (1989) categorized the diseases causesd by lentiviurses as either primary or secondary. Primary disease Is the major pathological manifestation of most lentivlral infections of domestic animals, and results from lentiviral replication in monocytes/macrophages, with dysregulation of cytokine production by these cells. Secondary disease results from opportunistic infections by other agents subsequent to loss of function of CD4+ T helper cells. Secondary disease has thus far only been demonstrated in infections of humans, non-human primates, and cats.
The fact that most retroviruses cause persistent, life-long infections in their hosts Implies an ability to avoid elimination by host immune responses. An important strategy used by retroviruses to avoid elimination is suppression of host immunity. As reviewed by Specter et al. (1988), both humoral and cell-mediated host responses are affected by retroviral infections of mammals. At least eight different murine retroviruses have been reported to inhibit primary and/or secondary antibody production. Both feline leukemia virus (FeLV) and FIV are also known to inhibit antibody responses to newly encountered antigens, as does HIV. Cell-mediated responses may also be severely depressed, particulariy in HIV and FIV infection. In addition, functions of cells of the innate immune system, such as macrophages and NK cells, are often adversely affected. In fact, it has been suggested that the T cell dysfunction seen in human HIV infection is predominantly due to defective antigen presentation (Meyaard et al., 1993). In general, 6
retroviaises can suppress host immune functions by replicating in immune cells, with direct cytopathic effects, by activation of cells with suppressor activity, or by the production of suppressive products such as p15E of FeLV (Specter et al., 1988). In the case of HIV infection, many other mechanisms may play a role, such as the induction of programmed cell death, autoimmunity, or overwhelming antigenic diversity (Weiss, 1993). Thus, in addition to being Important pathogens of humans and domestic animals, the retroviruses have provided an impetus to increase our understanding of interactions between pathogens and the host immune system.
Discovery of bovine retroviruses
In the late 1960's and early 1970's, three bovine retroviruses were isolated during the search for a viral cause of the disease of cattle known as enzootic bovine leukosis. A significant breakthrough occurred with the demonstration that lymphocytes from some lymphosarcomatous cattle produced virus particles after being cultured in the presence of mitogens (Miller et al., 1969). Subsequent epidemiological studies and transmission experiments showed that this virus was associated with both tumor formation and persistent lymphocytosis, and it was named bovine leukemia virus. At about the same time, bovine syncytial virus (BSV), a spumavirus, was isolated by Malmquist et al. (1969) from both lymphosarcomatous and nomial cattle. It has not yet been associated with any significant bovine health problems (Jacobs et al., 1995). A third bovine retrovirus was isolated from the buffy coat leukocytes of an eight year old Holstein cow (R29) with persistent lymphocytosis, progressive weakness, and emaciation (Van Der Maaten et al., 1972). After further genetic and immunological characterization revealed it to be a lentivirus (Gonda et al., 1987), this virus was named bovine 7 immunodeficiency-like virus due to similarities to HIV. Most recently, the cause of Jembrana disease, an acute and often fatal infectious disease in Bali cattle (Bos javanicus) in Indonesia has been identified as a lentivirus that is closely related to BIV (Chadwick et al., 1995). Interestingly, the disease caused by this virus may be very species-specific, since inoculation of European cattle did not produce clinical signs (Soeharsono et al., 1990). Because the Jembrana disease virus has not been shown to occur in the United States, the remainder of this review will focus on BIV and BLV infection.
Bovine Immunodeficiency Virus (BIV)
BIV genetics and replication
The genome of BIV is the most complex of the nonprimate lentiviruses (Fong et al., 1995). In addition to the obligate retroviral structural genes gag, pol, and env, BIV contains at least five open reading frames (ORFs) in a "central region" between and overlapping the pol and env genes (Gonda et al., 1987, Garvey et al., 1990). Three of these were identified as tat, rev, and v^by their similarity to regulatory genes of other lentiviruses, while two unique ORFs of unknown function were designated IVand /(Fong et al., 1995). Similar to EIAV and visna virus, BIV apparently lacks a nefgene (Garvey et al., 1990), which is thought to function in HIV as a transcriptional silencer important in down-regulating viral expression, (Cullen, 1992). The proviral long terminal repeat (LTR) of BIV serves as the viral promoter for transcription initiation. The BIV tat gene codes for a transactivating protein. Tat, which has been shown to increase viral transcription from the 5' LTR 15 to 80 fold 8
(Pallansch et al., 1992). BIV Tat interacts with a post-transcriptional regulatory element, TAR, in the nascent viral RNA (Carpenter et al., 1993). Unlike the HIV Tat protein, BIV Tat apparently does not use cellular transcription factors to stabilize binding to RNA in vivo (Chen and Frankel, 1994). However, a relatively high level of BIV LTR activity was observed in the absence of BIV Tat in cells transfected with a BIV LTR-reporter gene construct, suggesting that cellular factors are also involved in regulating BIV gene expression (Pallansch et al., 1992). In support of this, several cis-acting regulatory elements have been identified in the BIV LTR which may play a role in determining the extent of gene expression in different ceil types (Fong et al., 1995). The Rev protein of lentiviruses promotes the stability and export of unspliced and single spliced mRNA coding for Gag and Env structural proteins, respectively (Cullen, 1992). The timing of Rev expression therefore plays an important part in the shift from early to late replication in complex retroviruses. The BIV Rev protein has been detected in BlV-infected cells, mainly in the nucleoli, and its expression was correlated with the appearance of Gag and Env proteins (Oberste et al., 1993). Genes coding for BIV structural proteins have also been characterized. The BIV gag gene codes for the Gag precursor, p53, which is processed by a viral protease into matrix (pi6), capsid (p26), and nucleocapsid (pi3) proteins (Battles et al., 1992), as well as 3 smaller fragments (Tobin et al., 1994). Several alternative Gag cleavage products have also been identified using monospecific sera raised against Gag peptides (Battles et al., 1992). The N terminus of the matrix protein of BIV was shown to lack the modification by fatty acids that is found in most lentiviruses (Tobin et al., 1994). Since this post-translational modification by fatty acids is thought to be important In localizing the Gag precursor of some retroviruses to the plasma membrane during vims assembly, BIV may use an 9
alternative mechanism. The BIV envgene codes for a 145 kDa precursor protein which is cleaved into the surface envelope gp110 and the transmembrane protein gp42 (Rasmussen et al., 1992). Genomic variability is a well-known feature of lentiviral infections, and BIV is no exception. Even though they were both derived from a single virus inoculation, the BIV 106 2ind BIV 127 infectious molecular clones were shown to have considerable nucleotide and amino acid sequence variation, particulariy in the surface envelope genes (Braun et al., 1988, Garvey et al., 1990). Similarly, more recent studies have shown that BIV isolates from Florida and Oklahoma have larger surface envelope genes compared to these BIV R29-derived molecular clones (Suarez et al., 1995). These findings lend support to the concept that a lentivirus such as BIV is actually a quasi-species consisting of many variant genomes.
Host immune responses to BIV infection
Both T cell mediated and antibody responses can be important components of the host response to viral infections. However, virtually no information is available about cell-mediated responses to BIV infection. Recombinant BIV Gag and Env proteins expressed in E. coli have been used to determine major epitopes recognized by antibodies in the serum of BlV-infected cattle; one in the carboxy temiinus of BIV p26 (Atkinson et al., 1992), and another in the amino temiinus of BIV gp42 (Chen et al., 1994). Antibody responses to specific BIV proteins have been characterized in BlV-infected cattle by Western immunoblotting using pelleted virus preparations from BlV-infected cell cultures (Whetstone et al., 1990, 1991). Antibodies to the BIV capsid protein, p26, were detected as eariy as 2 weeks following BIV inoculation. Reactivity to the surface envelope glycoprotein gp110 was detectable in cattle by 4 to 6 weeks p.i. It was also reported that antibodies to p26 persisted for more than two years p.i. in experimentally infected cattle, depending on the sensitivity of the assay used (Whetstone et al., iggo). However, data from this study indicated that two out of the four animals tested at 19-38 months p.i. had lost detectable antibody, as determined by Westem blotting and immunofluorescence (IFA). Moreover, in one BlV-infected calf, predominant antibody reactivity shifted from core to envelope several months post-BIV inoculation (Whetstone et al., 1991). In a more recent study, although sera from eight BlV-infected cattle contained antibodies reactive with a BIV p26-containing recombinant protein by four weeks post-inoculation (p.i.), seven animals lost detectable p26-specific antibodies by six months p.i. (Isaacson et al., 1995). Therefore, serological assays which rely on the detection of antibody reactivity to the major BIV core protein may underestimate the true prevalence of BIV infection.
Prevalence of BIV infection
The prevalence of bovine retroviral infections has largely been estimated by serologic surveys for virus-specific antibodies. An inherent assumption of these methods is that all infected animals produce a detectable quantity of antibodies to viral antigens present in the assay. As noted in the previous section, this is a potential problem In the diagnosis of BIV Infection. Other assays based on the detection of BIV proviral DNA in BlV-infected lymphocytes by PGR have been developed (Nadin-Davis et al., 1993, Suarez et al., 1995b), but the labor-intensive nature of these tests render them impractical for screening large numbers of samples. The results of several serological surveys for antibodies to BIV are 11 summarized in Table 1. Studies of herds containing greater than 100 cattle found between 4 and 50% prevalence rates of BIV antibodies, while even greater variation was seen in studies of smaller herds.
Studies of BIV pathogenesis
Natural infections
As illustrated in Table 1, rates of natural BIV infection can vary considerably between different herds. Unfortunately, only a few seroepidemiological studieshave attempted to determine whether higher levels of BIV infection are associated with increased disease rates. Whetstone et al., (1990) studied four dairy cattle herds, two of which had persistent health problems. One of these latter herds had a high incidence of lymphoma and enlarged hemal lymph nodes; 69% of animals In this herd were BlV-positive by IFA. However, 74% were positive for
BLV, although there was no correlation between infection with the two viruses.
None of the 45 cattle In the other three herds had evidence of BIV antibodies. Thus it is difficult to draw conclusions from this study about the relationship of detectable
BIV antibodies and the development of disease. In a larger study, McNab et al.
(1994) detemnined that BIV seropositivity was significantly associated with lower milk production in dairy cows from 265 Canadian herds. These authors suggested that BlV-infected cattle that appeared clinically nomnal may have been predisposed to other subclinical conditions that affected production. In contrast, Detilleux et al.
(1995) showed that cows with the highest predicted transmitting ability for kilograms of milk fat plus protein had the highest prevalence of antibodies to BIV. 12
Table 1. Prevalence of BIV seropositivlty in the United States and Canada. Percentages of BIV seroprevalence for each study are indicated in parenthesis; data based on a single herd are noted. Abbreviations; IFA, immunofluorescence assay; ELISA; enzyme-linked immunosorbent assay; WB, westem immunoblotting.
BIV seropositive (%) Location Assay Reference
11 of 97 (11.3%) Louisiana IFA Amborski et al.,1989
6 of 138 (4.3%) Colorado IFA Amborski et al.,1989
80 of 1997 (4.0%) Southem U.S. IFA Black et al.,1989
54 of 78 (69%) not reported IFA,WB Whetstone et al.,1990 (single herd)
20 of 95 (21%) Colorado ELISA Cockerell et al.,1992
51 of 928 (5.5%) Ontario WB McNab et al.,1994
49 of 920 (5.5%) Ontario WB Jacobs et al.,1995
221 of 442 (50%) Mississippi Gag ELISA St. Cyr Coats et al., 1995
31 of 37 (84%) Louisiana ELISA Snider et al.,1996 (single herd)
Most recently, Snider et al. (1996) reported a high rate of BIV infection in Holstein cows from a single herd with a high rate of encephalitis, lymphoid cell depletion, reproductive problems, and secondary infections. However, the number of BlV-free cattle in this study was too low to allow a valid statistical comparison of the effect of BIV. In addition, there was a high rate of BLV infection in these cattle, and other potential causes of the chronic health problems of these animals were not investigated. While it is possible, as these authors suggested, that BIV infection had a profound effect on these cattle that were exposed to various environmental 13 stresses, further epidemiological studies will be required to delineate the role of BIV in bovine disease.
Experimental infections In order to reduce some of the inherent variability and extraneous factors that complicate the interpretation of studies of naturally infected cattle, the pathogenic effects of BIV infection have also been investigated in animals inoculated with BIV. Calves inoculated with infected cells or cell culture supematants from the original R29 cow developed enlargement and follicular hyperplasia of hemal lymph nodes, as well as a biphasic, transient lymphocytosis (Van Der Maaten et al., 1972). These studies were extended by Carpenter et al. (1992), who found that calves infected with BIV R29 for six weeks had transient fever, leukopenia, and lymphocytosis, as well as follicular hyperplasia in lymph nodes, hemal nodes, and spleen. These changes were similar to those observed in the early stages of HIV infection and after experimental infection of cats with FIV. However, because bovine viral diarrhea virus (BVDV) was present in the BIV inocula, it was unclear whether the clinical changes were due to BIV, BVDV, or both viruses (Carpenter et al., 1992). Two calves inoculated with BVDV-free Florida isolates of BIV developed a transient increase in mononuclear cell counts, but no fevers or enlarged lymph nodes were observed (Suarez et al., 1993). Lentiviral infections of humans, nonhuman primates, and cats have been associated with lymphomas (Cotter, 1993). Interestingly, an atypical T cell lymphosarcoma was reported in an 11 month old Holstein calf that had been experimentally infected with BIV for five months (Rovid et al., 1996). Although the role of BIV in tumor development was unclear, the animal had a monocytosis and 14 increased lymphocyte blastogenesis prior to tumor development, suggesting that BlV-induced immune stimulation might have contributed to tumorigenesis. Overall, despite some interesting observations, the role of BIV infection in diseases of cattle remains unclear. However, the asymptomatic stages of lentiviral infections in other species are often accompanied by altered immune functions. This suggests that BIV might similariy affect the immune responses of cattle in the absence of overt disease.
Effects of BIV infection on the host immune system
Effects on innate immunity Flaming et al. (1993) examined neutrophil functions in cattle experimentally infected with BIV. Although neutrophil functions in BlV-infected cattle at two months p.i. did not differ from those of age-matched controls, neutrophil antibody dependent cell mediated cytotoxicity (ADCC) was decreased in another group of cattle infected at four to five months p.i. In addition to decreased ADCC, neutrophil iodination was also decreased in a third group of cattle infected with various BIV preparations or infectious molecular clones for 19-27 months. Thus, BIV infection was shown to affect neutrophil functions, and duration of infection may have Influenced the nature of the effects on neutrophils. Monocyte functions have also been examined In cattle experimentally infected with BIV. Monocytes from three infected calves had decreases in superoxide anion release, phagocytic activity, and chemotactic responses compared to uninfected controls (Onuma et al., 1992). In contrast, monocyte functions were found to be nomnal in BlV-infected cattle through four years p.i. (Rovid et al., 1995). In this latter study, however, monocyte functions could be 15
inhibited in vitro by adding killed BIV or protein, suggesting that BIV is capable of inhibiting nnonocytes, but levels of virus replication In vivo may not have been high enough to inhibit the functions of circulating monocytes.
Effects on acquired immunity In infections with other lentiviruses such as HIV and FIV, it is well established that host immune functions are often suppressed prior to more obvious alterations in circulating lymphocyte numbers or development of clinical disease (Shearer et al., 1986, Miedema et al., 1988, Torten et al., 1991, Bishop et al., 1992). Despite ongoing conjecture about the potential of BIV to cause immunosuppression In cattle (Brownlie et al., 1994), as well as the potential value of BIV as a model for other lentiviruses such as HIV, there have been few controlled studies of the effects of BIV infection on acquired immune responses. Stott et al. (1989) reported that two calves inoculated with BIV R29 had decreased numbers of circulating CD4+ cells at three to four months p.i. However, because these results have only been published in abstract fomn, it is difficult to evaluate their significance. In several long-temi studies of a single group of cattle experimentally infected with BIV R29, no consistent differences have been seen in the circulating lymphocyte subpopulations of BlV-infected cattle compared to age matched controls (Flaming et al., 1993, Flaming et al., in press, Isaacson et al., in preparation). Problems in experimental design have also plagued some studies of lymphocyte functions in BlV-infected cattle. Lymphocyte proliferative responses to the mitogens concanavalin A (Con A), pokeweed mitogen (PWM), and phytohemagglutinin (PHA) were reported to be decreased in two calves six months p.i. (Martin et al., 1991), but these conclusions were reached by comparison to a single control calf, which could have had higher proliferative responses simply by 16 chance. In contrast, lymphocyte proliferative responses to PHA were shown to be significantly increased in two groups of cattle infected with BIV for either 4-5 months or 19-27 months (Flaming et al., 1993). However, a long-term study of the first group of cattle showed that lymphocyte proliferative responses to mitogens were consistently decreased in cattle infected with BIV compared to uninfected controls, although these differences were not statistically significant (Flaming et al., in press). During the fifth year p.i., the Con A-stimulated proliferative responses of lymphocytes from these same BlV-infected cattle were slightly, but significantly, suppressed (Isaacson et al, in preparation). Thus it appears that the effect of BIV on lymphocyte functions may be variable over the course of Infection. Despite the well-known ability of lentiviral infections such as HIV to suppress antibody responses (MIedema et al., 1988, Pinching et al, 1991), only a single report has been published on the effect of experimental BIV infection on antibody responses to non-BIV antigens (Onuma et al, 1992). These investigators reported a delayed humoral response to mouse serum proteins in one out of three BlV- infected calves compared to a single control calf. However, because of the low sample sizes involved, the delayed response in one calf could easily have been due to biological variation. In another study, there were no consistent differences in primary or secondary antibody responses to vaccination with tetanus toxoid (TT) in cattle infected with BIV for over three years compared to uninfected controls (Isaacson et al., in preparation). Therefore, there has been no clear evidence to date of suppression of acquired Immunity in controlled studies of BIV infected cattle. 17
Summary
BiV ts a lentivirus of cattle that is similar to immunosuppressive lentiviruses In terms of structure, genetics, and replication strategy. During the first few weeks after being inoculated with BIV, calves show relatively mild signs of lymphoproliferation. Later in the course of infection, some deficits in neutrophil function and lymphocyte responses to mitogens can be demonstrated in vitro, but there has been no unequivocal evidence of clinically apparent immunosuppressive disease in BlV-inoculated cattle. Moreover, the few well-controlled studies that have been reported to date have not shown any consistent, significant effects of BIV inoculation on acquired immune responses of cattle. Thus, although BIV remains a potentially important factor in bovine disease, there is clearly a need for further well-controlled studies of the effects of BIV infection on acquired immune responses in cattle.
Bovine Leukemia Virus (BLV)
General virology
Terminology Variations in the terms used to describe different manifestations of BLV infection has resulted in some confusion. The tenri bovine leukemia virus is itself misleading since leukemia implies an increase in circulating leukocytes due to invasion of primary tumor cells into the bloodstream, which is an uncommon event in BLV-induced neoplasms of cattle (Jacobs, 1980). A more descriptive term is bovine leukosis, since leukosis refers to a proliferation of the leukocyte-forming 18
tissue, and encompasses the lymphoid tumors as well as persistent lymphocytosis. The use of temis such as "leukotic" and "aleukemic" to describe cattle with various manifestations of BLV infection appears to mean different things to different authors, and should probably be avoided.
Structure and classification Oncoviruses were originally classified into types A through D based on their morphology and patterns of budding from the cell membrane. Although BLV is most similar in size and morphology to the C-type retroviruses of other species, the budding form of the virus is slightly different from the classic type C virus (Weiland and Ueberschar, 1976). Moreover, the reverse triansop|i(B8^aBliBt^irBqaiTtadf1^ other C-type viruses in which DNA synthesis proceeds equally well in the presence e oncogenioftelttraiilAn^, BLV is most closely related to the human T cell leukemia viruses (HTLV-I and -II), and the BLV/HTLV group is currently classified as a separate genus (Coffin, 1992). Infections with BLV or the HTLVs are characterized by a long latency period, relative absence of viremia, and lack of an oncogene or preferred proviral integration sites in their respective hosts.
Genetics and replication The complete nucleotide sequence of the BLV genome has been reported (Sagata et al., 1985). The integrated BLV provirus contains typical retroviral LTRs, along with gag, pol, and env structural genes. The BLV gag gene codes for a precursor polypeptide which is cleaved into proteins plO, pi 2, pi 5, and the major core protein p24 (Bumy et al., 1980). Similariy, the BLV env gene codes for a glycosylated 72 kDa envelope precursor which is cleaved into two envelope 19 proteins, transmembrane gp30 and surface gp51. The po/gene encodes the reverse transcriptase as well as a protease and integrase (Bumy et al., 1980). Like other complex retroviruses, BLV has several regulatory genes. The BLV rex gene codes for a regulatory protein which sen/es an analogous function to lentiviral Rev proteins; that is, it facilitates transport of mRNAs coding for structural proteins to the cytoplasm (Derse et al., 1988). Like the HTLVs, but unlike other RNA tumor viruses, BLV contains a pX region between the en/gene and the 3' LTR that contains open reading frames that code for regulatory proteins (Rosen et al., 1985, Derse et al., 1987). One of these, the 38 kDa BLV Tax protein, was shown to specifically activate viral transcription by interacting with proviral LTR sequences: however, there was no evidence that BLV Tax binds directly to DNA (Derse et al., 1987). Thus, Tax may increase viral transcription through interactions with cellular transcription factors that recognize enhancer elements within the BLV LTR. The HTLV/BLV group of retroviruses exhibits relatively little genetic variation compared to the lentiviruses. Viral subtypes have been Identified at the genetic level, but these do not appear to result in antigenic differences (Bumy et al. 1980). Interestingly, comparison of the nucleotide sequence of an Australian BLV isolate with isolates from other countries showed more variation at the amino acid level of the gag gene than in the env gene, but antigenic differences were not detennined (Coulston et al., 1990). The genetic variation of BLV in vivo has also been studied following infection of animals with infectious BLV clones. Isolation of BLV proviral DNA from the PBMC of sheep18 months after injection with cloned BLV provirus showed an in vivo rate of nucleotide change in the env gene of 0.009%, corresponding to only one amino acid change per year (Rovnak et al., 1993). Similarly, 20 independent BLV clones derived from the PBMC of a sheep that 20 developed lymphosaracoma after being inoculated with cloned BLV had mutation rates in the envand LTR genes of only 0.043 and 0.041 percent, respectively, which was close to the the error rate of Taq polymerase used to amplify the genes (Willems et al., 1995). This lack of genetic variation is attributable, at least in part, to the fact that the reverse transcriptase of the HTLV/BLV group is less error-prone than that of other retroviruses (Mansky and Temin, 1994). In addition, members of the HTLV/BLV genus are thought to replicate more often as proviruses during cell division, rather than as viruses by reverse transcription (Cann and Chen, 1990). Regardless of the mechanistic explanation, these findings suggest that antigenic variation of BLV may not be an important strategy to escape Immune recognition.
Clinical manifestations of BLV infection
Lymphoid tumors Between 0.1 and 10 percent of BLV-infected cattle develop malignant lymphoma or lymphosarcoma (Bumy et al., 1988). These tumors are seen most frequently in cattle between 5 and 8 years of age and often occur in familial aggregations (Ferrer et al., 1980). They are characterized by a proliferation of neoplastic lymphocytes that results in the formation of solid tumor masses and/or diffuse infiltration of neoplastic cells into various tissues. The lymph nodes are almost invariably involved, with tumors often spreading to the abomasum, heart, uterus, or other organs (Miller, 1981). Common clinical signs include anorexia, weight loss, agalactia, lymphadenopathy, and posterior paresis or paralysis (Johnson and Kaneene, 1991). Death almost always occurs within weeks to months of the onset of clinical signs. However, cattle with heart involvement may die suddenly with no previous signs of disease. Economic losses associated with 21
BLV infection of cattle in the U.S. have been recently estimated to be 44 million dollars annually due to reduced foreign trade and losses due to lymphoma (Thurmond, 1987).
Persistent lymphocytosis About 30 percent of BLV-infected cattle eventually develop persistent lymphocytosis (PL), which is generally defined as an abnormal elevation in numbers of circulating lymphocytes over a period of at least three months (Ferrer et al., 1979). Since increasing age is associated with decreasing lymphocyte counts in cattle, age is an important consideration in defining abnormal lymphocyte counts (Perman et al., 1970, Fen-er et al., 1979, Thurmond et al., 1990). Although PL may appear to be a pre-cancerous condition, most lymphocytotic cattle never develop tumors, and about one-third of the cattle that develop lymphosarcoma never progress through a PL stage. Thus PL is considered to result from a BLV-induced lymphoproliferatlon that is neither a subclinical form of cancer nor a prerequisite for tumor development. However, PL may not be a completely benign condition. Milk and fat production have been shown to decline in Holstein cattle with PL compared to uninfected animals (Wu et al.,1989. Da et al., 1993), and infected cows are culled from herds prematurely connpared to seronegative cattle (Brenner et al., 1989, Pollari et al., 1993). The factors that influence the development of the B cell proliferation that characterizes PL remain poorly understood. However, genetics cleariy play an important role. The tendency for BLV-infected cattle to develop PL has been shown to be linked to genetic factors such as major histocompatibility complex (MHC) or BoLA type (Lewin and Bemoco, 1986, Lewin et al., 1988). Recently, Lewin et al. (1994) showed that resistance to PL development was associated with 22 the presence of the amino acids Giu-Arg at the putative antigen binding region of the BoLA-DFRBS MHC class II molecule, suggesting a potential role for cell- mediated immune responses in confemng resistance to PL (Lewin et al., 1994).
BLV+, non-PL At least 60% of BLV-infected cattle do not develop tumors or PL (Ferrer et al., 1979), but the effect of BLV infection on this large group of cattle is still largely unknown. One difficulty is that most studies of the effects of BLV infection have not differentiated between PL and non-PL infected herdmates. Another problem is that many BLV-infected cattle may be culled from the herd prior to the development of clinically obvious health problems, thus the consequences of BLV Infection may be underestimated. In some early studies no significant differences were found between BLV-infected and uninfected cattle in production or reproductive perfomnance (Huber et al., 1981b). Interestingly, cows with a high genetic potential for milk production were more likely to be seropositive for BLV compared to herdmates with lower potential (Wu et al., 1989, Detilleux et al., 1995). Although the reason that BLV infection was associated with higher genetic potential is unknown, one possibility is that the increased physiological stress on higher- producing cattle might increase susceptibility to BLV infection (Wu et al., 1989).
Host immune responses to BLV infection
Humoral responses Virtually all BLV-infected cattle produce virus-specific antibodies (Ferrer et al., 1977). Miller and Van der Maaten (1976) used agar gel immunodiffusion (AGID) and complement fixation (CF) tests to study humoral responses to BLV 23 antigens in BLV-inoculated calves. BLV-specific antibodies were detected by both methods in all Infected calves within two to three months p.i. Although CF titers remained relatively stable throughout the first year of observation, many sera had declining levels of reactivity in the p24 AGID test, suggesting that the persistent reactions in the CF test were due to the presence of another BLV antigen, which was later determined to be gp51. The gp51 antigen was later incorporated into the AGID test which is intemationally recommended for screening cattle for BLV infection. Antibodies to gp51 are believed to be responsible for the BLV-neutralizing activity found in the serum of infected cattle (Van Der Maaten and Miller, 1990). Monoclonal antibodies have been generated against gp51 and used to identify eight antigenic sites (named A through H) on the gp51 molecule (Bruck et al., 1982a). Sites F, G, and H were shown to be neutralizing epitopes involved in syncytia formation, and site G appeared to be a target for antibody mediated complement-dependent cytotoxicity (Bruck et al., 1982b). Once infection is established, however, the presence of antibodies to BLV gp51 or other antigens does not appear to hinder the development of PL or lymphoma, since animals with lymphosarcoma usually have higher antibody titers against BLV antigens (Deshayes et al., 1980, Heeney et al., 1988).
Cell-mediated responses T cell-mediated responses may play a critical role in protection against BLV infection and/or disease. Inoculation of sheep with a vaccinia virus expressing BLV gp51 strongly suppressed titers of BLV following challenge (Ohishi et al., 1991). Antibodies to gp51 were not detected in vaccinated sheep, but strong delayed-type hypersensitivity responses developed against gp51 (Okada et al., 1993). 24
Therefore, both humoral and cell-mediated immune responses appear to contribute to host resistance against BLV infection. Gatei et al. (1993) measured CD4+ and CD8+ T cell responses from cattle and sheep to a complete series of overlapping peptides from gp51, each 20 amino acids In length. Protection of sheep against BLV challenge was correlated with CD4+ T cell responses to gp51 peptide 51-70. Cytotoxic T cell responses were generated only against peptides 121-140 and 131-150. This information should be useful in ongoing efforts to develop an effective BLV vaccine.
Detection and prevalence of BLV infection
Prior to the discovery of BLV, the presence of PL was used as an indicator of cattle most likely to develop bovine leukosis (Bendixen, 1965). This method was unsatisfactory, however, because some infected animals never become lymphocytotic while others develop tumors without progressing through a PL stage. Once the virus was identified, methods were developed to detect BLV In cultured peripheral blood mononuclear cells (PBMC). These included direct observation by electron microscopy (Miller et al., 1969), induction of syncytia on susceptible cells (Ferrer and Diglio, 1976), or immunological detection of BLV p24 expression (Jerabek et al., 1979). Though these methods tend to be quite sensitive, most require short-term culturing of PBMC of infected animals to induce virus expression. Cell culture is not required for methods that are based on the detection of BLV proviral DNA in infected cells. One such method is a highly sensitive polymerase chain reaction-based test which allows the detection of as few as 10 copies of BLV in BLV-infected PBMC (Sherman et al., 1992). 25
Assays that have been developed to detect BLV-specific antibodies in bovine serum include the AGID (Miller et al., 1972), complement fixation (Miller and Van Der Maaten, 1974), immunofluorescence using fixed BLV-infected cell cultures (Ferrer et al., 1975), radioimmunoassay (RIA) (McDonald and Fen'er, 1976), and radioimmunoprecipitation (Deshayes et al., 1980). A comparison of four serological tests for the detection of BLV-specific antibodies showed that the greatest number of positive sera was detected by an RIA for gp51 (188), followed by RIA for p24 (187), virus neutralization (183), and AGID for gp51 (176) (Miller et al., 1981). Thus, although false-negative test results are more of a problem with the AGID test, there was 96.8% agreement between the tests. Most recently, an ELISA for detection of antibodies to BIV gp51 has been made commercially available. Rates of BLV infection in U.S. cattle have most commonly been estimated using the gp51 AGID test. Rates of BLV infection in U.S. herds are generally higher than those reported for BIV infection, ranging from 24% to 66% in dairy cattle, with lower rates found in beef herds (Johnson and Kaneene, 1991). As with BIV, seroprevalence rates as high as 95% have been reported for individual dairy herds (Huber et al., 1981 a). A study of 3000 Michigan dairy cattle over two years of age in 82 herds revealed prevalence rates between 24 and 36% by AGID (Kaneene et al., 1983). Longitudinal studies have also shown that BLV infection is more common in cattle over 2 years of age, which may be related to the closer physical contact and more intense management practices that cattle are exposed to at this time (Johnson and Kaneene, 1991). Seroepidemiological studies have also revealed a genetic influence on susceptibility to BLV infection. Cattle with BoLA type Eu28A had a significantly higher incidence of antibodies to BLV than those with type W10 (Stear et al.,1988) or W8.1 (Lewin et al., 1988a). Since the Eu28A allele is very common, these 26
authors speculated that cattle sharing the same MHC type might be less likely to destroy BLV-infected foreign cells, thus transmitting the infection more effectively. However, since other MHC alleles would be expected to differ in outbred cattle, the reason for this association remains uncertain.
Pathogenic mechanisms in BLV infection
BLV target cells There is considerable evidence that BLV replicates In B cells. When B cells of BLV-infected cattle were separated from PBMC by nylon wool adherence and cultured, 39% of the B-cell enriched preparation produced BLV antigen in vitro compared to 0.5% of the non-B cells (Paul et al., 1977). Kenyon and Piper (197Tb) showed that only PBMC or B ceils purified by various means would produce syncytia on bovine embryonic spleen cells. The production of BLV gp51 protein was detected by immunofluorescence in B cells isolated from the PBMC of four naturally infected lymphocytotic cattle either by T cell depletion or by flourescence activated cell sorting (Jensen et al,, 1990). Although B cells are the predominant target of BLV infection, it appears that non-B cells may also be infected under some conditions. Integrated proviral BLV DNA was detected in mitogen-stimulated CD4+ and CD8+ T cells from seven out of ten lymphocytotic cattle, although this could have been the result of in vitro infection (Stott et al, 1991). Similariy, Williams et al. (1988b) reported the presence of BLV proviral DNA in immunoaffinity purified T cells from BLV+, non-PL cattle. B cells from a lymphocytotic cow had the highest BLV load as determined by PCR, but BLV proviral DNA was also found in highly purified CD8+ T cells, monocytes, and 27 granulocytes (Schwartz et al., 1994). These findings suggest the BLV receptor is not restricted to a specific cell population. A cDNA coding for a polypeptide with properties of a putative BLV receptor has been isolated from MDBK cells and characterized (Ban et al., 1993). BLV gp51 bound to recombinant phages expressing the cDNA, and mammalian cells transfected with the cDNA became susceptible to BLV infection. The protein was predicted to have a receptor-like structure with a gp51 binding region, a single transmembrane region, and a putative intracellular cytoplasmic domain. The cDNA hybridized with DNA from a wide variety of species, coinciding with the known broad host range of BLV.
Tumorigenesis A large amount of research effort has been devoted to mechanisms of BLV tumorigenesis, and several published reviews have dealt extensively with this subject (Bumy et I., 1988, Schwartz and Levy, 1994). Therefore, only those aspects of BLV tumorigenesis with particular relevance to overall BLV pathogenesis will be summarized here. The BLV provirus is clonally integrated in tumors from individual cattle, but the provirus is integrated at different sites in different animals, providing evidence against the possibility that cell transformation results from cis-activation of host proto-oncogenes (Kettmann et al., 1980, 1985). A study of cells from lymphoid tumors from 17 cattle seropositive for BLV revealed that all were B cells containing one to four copies of BLV proviral DNA (Heeney and Valli, 1990). By comparison, proviral integration was shown to occur at a large number of chromosomal sites In 25-33% of circulating lymphocytes of lymphocytotic cattle (Kettmann et al., 1980) 28
Studies of cell clones derived from BLV-induced tumors have shown that all contain at least a portion of the BLV provirus, with the 3' region always being present (Kettmann et al., 1985). However, the expression of viral RNA is not required for the maintenance of the transformed state. Thus, it appears that while the 3' region of the BLV provirus is essential for the induction of neoplasia, once the required cellular genes are activated, the continued expression of the BLV proteins becomes unnecessary. Since cells expressing viral antigens are more likely to be destroyed by the anti-BLV immune response, there may be immune selection in vivo against transformed cells that continue to express viral proteins (Kettmann et al,, 1985). Another possible mechanism was suggested by the finding of a rare Tax cDNA in uncultured lymphocytes from BLV-infected cattle (Jensen et al., 1991). This showed that the Tax mRNA could itself be reverse transcribed and possibly integrated back into the host genome. If transcription of the integrated tax gene was controlled by a host cell promoter, leukemogenesis or lymphocyte proliferation could be the result.
BLV expression in vivo Because it is difficult to detect BLV mRNA, proteins, or virions in freshly Isolated lymphocytes from BLV-infected cattle, BLV expression has long been thought to be restricted at the transcriptional level in vivo. (Baliga and Ferrer, 1977, Kettman et al., 1980). However, the persistence of a strong humoral response to several BLV proteins in neariy all infected cattle implies that viral antigens are expressed at sufficient levels in vivo to sustain a host immune response. Low levels of BLV-specific mRNA have recently been demonstrated in freshly isolated PBMC from BLV-infected cattle by highly sensitive techniques such as in situ hybridization (Lagarias and Radke, 1989) and polymerase chain reaction (PGR) 29
(Jensen et al., 1991). BLV RNA has also been detected In the plasma of a BLV- infected cow by reverse transcriptase-directed PGR using probes specific for BLV pX or pol (Sherman et al., 1992). There appear to be different patterns of BLV expression in PL vs. non-PL cattle. Using RT PGR, the fax/rex message was detected in the PBMC of five out of five lymphocytotic cattle, compared to one out of seven infected cattle that were hematologically normal (Gockerell and Rovnak, 1988). This correlates with the greater number of Infected cells In PL animals (30%) versus non-PL (< 5%) (Kettmann 1980), suggesting that there may be fundamental differences in the regulation or immunologic control of BLV expression in PL vs. non-PL cattle. In addition to the fax/rex mRNA, other small mRNAs have been shown to be produced from the pX region by altemative splicing (Alexandersen et al., 1993). One of these, the GIV mRNA, was detected in the PBMC of both naturally and experimentally infected cattle, and the highest level of GIV mRNA was found in the calf with the highest lymphocyte counts. This suggests that the GIV product may be involved in stimulating lymphocyte proliferation. A conserved seven amino acid sequence in the GIV protein had identity to the myb family of proto-oncogehes, suggesting a possible role in regulation of cellular or viral transcription. Although low levels of BLV mRNAs encoding regulatory proteins have been detected in freshly isolated PBMC from infected animals, the expression of mRNA coding for structural proteins appears extremely restricted. Therefore, it is unclear whether persistent antibody responses to BLV structural proteins in infected cattle result from continuous expression of these proteins at a very low level, or periiaps intermittent expression during periods of lymphocyte activation. An altemative hypothesis is that other host tissues besides PBMC may provide a source of BLV antigens in vivo. To investigate this possibility, the expression of BLV antigens was 30 examined in lymph node sections from BLV-infected cattle. Although cells expressing BLV antigen were present in intrafollicuiar and marginal zones, they were infrequent and solitary (Heeney et al., 1992). In another study, three persistently infected primary cell cultures derived from adherent ceils in PBMC of BLV-infected PL cattle were identified as endothelial cells (Rovnak et al., 1991). These cells had characteristics of lymph node post-capillary endothelial cells. Because similar cells are thought to come in intimate contact with lymphocytes that are migrating into the lymph nodes from the blood, this may be a site where lymphocytes could be infected in the absence of viremia.
Plasma blocking factor The expression of BLV by cultured PBMC from BLV-infected cattle can be blocked at the transcriptional level by a soluble factor present in bovine plasma (Gupta and Ferrer 1982, Gupta et al., 1984). This plasma BLV blocking factor (PBB) is a non-immunoglobulin, non-interferon protein molecule with an apparent molecular weight of 150 kDa (Gupta et al., 1984). It is most commonly present in the plasma and lymphatic fluid of BLV-infected cattle, but it is also found in some BLV-free cattle (Gupta et al., 1989, Zandomeni et al., 1992). The blocking effect is reversible, since BLV production resumes after PBB is removed and lymphocytes are washed (Gupta and Ferrer, 1982). In addition, a platelet-derived factor capable of suppressing BLV PBB has been described (Tsukiyama, 1985). Other investigators have suggested that anti-BLV antibodies, especially those directed against gp51, may contribute to BLV-suppression in vitro (Taylor et al., 1993). Although the effect of these factors on BLV transcription has only been demonstrated In vitro, they very likely contribute to the repression of BLV transcription in vivo. 31
Studies of BLV replication in vitro Although BLV expression appears to be tightly regulated in vivo, PBMC from BLV-infected cattle produce viral mRNA, proteins, and virions within only a few hours after being placed in culture (Miller et al., 1969, Baliga and Ferrer, 1977, Driscoll et a!., 1977, Jensen et al., 1990). This differs from observations of other oncogenic retroviruses such as feline leukemia virus, which replicates continuously in lymphocytes in vivo or in vitro (Baliga and Ferrer, 1977). Thus BLV is an Interesting virus for the study of factors that control the expresssion of a leukemia virus in lymphoid cells. During antibody responses to T cell-dependent antigens in vivo, T cells stimulate B cell proliferation and differentiation by expressing costimulatory molecules on their surface which interact with ligands on B cells, as well as by secreting soluble cytokines. The influence of T cells and T cell-derived factors on the expression of BLV by infected B cells has been investigated. It has long been known that PHA, a T cell mitogen, enhances BLV expression in cultured lymphocytes (Muscoplatt 1974b, Baliga and Ferrer, 1977), and conditioned medium from cultured T cells of BLV-free sheep greatly enhanced BLV production by infected B cells (Djilali et al., 1987). However, only a slight reduction in BLV expression was observed in cultures of isolated ovine B cells compared to nonfractionated PBMC (Jensen et al., 1990). Therefore, while T cell-derived factors appear to enhance BLV expression, T cells do not appear to be required for BLV expression In vitro. Because BLV expression is augmented in vitro by factors that activate lymphocytes such as mitogens oranti-IgM antibodies (Lagarias and Radke, 1989), it has been hypothesized that activation of BLV-infected cells In vivo may stimulate BLV replication. This possibility was investigated in vitro using BLV-infected fetal 32 lamb kidney (FLK) cells synchronized with respect to cell cycle stage. The production of BLV antigens was found to be increased during the S and G2 phases and decreased during M and G1 (Takashlma and Olsen, 1981). Treatment of FLK cells with DNA synthesis inhibitors hydroxyurea and mitomycin C inhibited BLV antigen production, thus It was suggested that the production of BLV antigens requires that infected lymphocytes enter the S phase of the cell cycle. However, Chatterjee et al. (1985) found that the enhancing effect of PHA on BLV expression in vitro was not due to the mitogenic activity of PHA, since PHA stimulation of BLV antigen production by cultured lymphocytes occured prior to any detectable Increase In either DNA or RNA synthesis. The PHA effect was abolished by inhibition of protein synthesis with cyclohexamide, indicating that the enhancement of BLV expression was mediated by a PHA-induced protein. In general, various studies have shown that BLV expression in vitro is enhanced by several agents which are known to activate B cells. As noted previously, BLV transcription has been shown to be regulated by the Tax protein, which may enhance the binding of cellular proteins to the viral LTR. One host factor that has been shown to be Involved In BLV transcription is the cyclic AMP- responslve element (CREB) protein. Three cyclic AMP response elements have been found upstream of the transcriptional start site in the BLV LTR (Willems et al., 1992). Nuclear lysates of B cells from BLV-infected lymphocytotic cattle contained proteins that specifically bound to the Tax responsive element In the BLV LTR (Adam et al., 1994). A monoclonal antibody against the CREB protein supershifted a protein-DNA complex present only in BLV-infected B cells, indicating that the CREB protein is involved in BLV transcription in vivo. In addition, an NF-kB binding site has been identified upstream of the transcriptional start site in the BLV LTR (Brooks et al., 1995). This site was shown to bind several members of the kB family 33
of proteins, and to function as an enhancer element when inserted upstream of a reporter gene. Since NF-kB is constitutively expressed in B cells (Mauxion et al., 1991), interactions between NF-kB and the BLV LTR may be responsible for a low basal level of BLV transcription in vivo. Upon B cell activation, an increased binding of NF-kB to the BLV LTR could further enhance BLV replication.
Spontaneous lymphocyte proliferation (SLP) Cultured lymphocytes from BLV-infected animals with or without PL often display high levels of spontaneous 3H-thymidine uptake in the absence of mitogenic stimulation in vitro (Muscoplat et al., 1974b, Kenyon and Piper 1977a, Takashima and Olsen, 1980, Williams et al., 1988b, Rovnak et al., 1993, Sordillo et al., 1994). Although the mechanistic basis for SLP is unclear, the timing of SLP coincides with BLV expression in vitro (Rovnak et al., 1993). A positive correlation was shown between a BLV-infected animal's circulating lymphocyte count and the amount of ^H-thymidine incorporated by cultured PBMC during the first hour after blood sampling (Chandler and Gilman, 1974). Since a greater percentage of lymphocytes are infected in lymphocytotic cattle (Kettmann et al., 1980, Cockerell and Rovnak, 1988), the degree of SLP may be detemnined by the virus load of a particular animal. Two main explanations have been offered to account for SLP; 1) that it results from a specific immune response to BLV antigens produced in vitro, and; 2) that it reflects a polyclonal lymphocyte proliferation induced by BLV or BLV proteins. In an eariy study, Kenyon and Piper (1977b) determined that SLP resulted from expansion of an uninfected B cell fraction, and they suggested that this was a specific response to BLV antigens. This hypothesis was strengthened by the observation that anti-BLV serum added to lymphocyte cultures abrogated 34 the proliferative response by a non-cytolytic mechanism (Takashima and Olsen, 1980, Thom et a!., 1981). Thus the available evidence appears to support the Idea that SLP in BLV-lnfected cattle results from a virus-specific immune response. In the HTLV system, however, it has been shown that soluble HTLV Tax protein added to cultured human lymphocytes rapidly translocates to the nucleus and induces SLP as well as increasing the expression of the interleukln-2 receptor alpha chain (Mariott et al. 1991, 1992). The use of fax expression plasmids would allow analogous experiments to be performed with bovine cells, and would facilitate the generation of BLV Tax-specific monoclonal antibodies for use in further investigations of the role of Tax in BLV pathogenesis.
Effects of BLV infection on the host immune system
B cell numbers By definition, BLV-infected cattle with PL have elevated numbers of circulating lymphocytes in their peripheral blood. Several methodologies have been used to Identify the affected lymphocyte subsets. Prior to the development of monoclonal antibodies (MAbs) specific for lymphocyte surface markers, T cells could be identified by their tendency to form rosettes when incubated with sheep red blood cells (sRBC), and B cells by rosette fomnation with complement-coated sRBC. The development of well-characterized and workshop-confirmed MAbs against bovine B and T cell markers has allowed a more detailed characterization of changes that occur in circulating lymphocyte subsets of BLV-infected cattle using either immunofluorescence techniques or flow cytometry. The findings of many of these studies are summarized in Table 2. 35
Table 2. Reported B ceil percentages and/or absolute numbers in control vs. BLV- Infected cattle with or without PL. A single number indicates that only group means were reported for a particular study. Absolute numbers of cells (10^ cells/^l) are listed for those studies that reported them. The number of cattle in each group is given in parentheses, 'Reported to be significantly different from the control group (P < 0.05).
Control (n) BLV+, non-PL (n) BLV+, PL (n) Reference
28% (12) ND 63% (4) Muscoplat et al., 1974a
15-30% (14) 12-50% (8) 54-76% (6) Kenyon and Piper, 1977a
15-30% (6) ND 28-58% (8) Takashima et al., 1977
16% (12) ND 62%* (5) Kumar et al., 1978
11% (20) ND 59% (3) Paul et al., 1979
29% (5) 40% (11) 85% (10) Esteban et al., 1985
18% (12) 16-58% (2) 48-82% (4) Possum et al., 1988
26% (4) 21% (4) 58%* (3) Williams et al., 1988a
37% / 4002 (67) 45%/4144 (125) 78% / 15645 (16) Lewin et al., 1988b
30%/1880 (6) 29% /1429 (6) 75% / 8785* (6) Gate! et al., 1989
31%/1681 (15) 29% /1095 (16) 73%/9911* (11) Taylor et al., 1992
1370 (4) 691 (4) 13633 (4) Meirom et al., 1993
1240 (7) 2010 (7) 9820 (2)* Sordillo et al., 1994
498 (6) ND 8666 (8)* Stone et al., 1994
16%/560 (6) ND 57% / 8593 (8)* Stone et al., 1995 36
Virtually all these studies found that BLV-infected cattle with PL have a drannatic increase in the percentage and/or absolute number of circulating B cells, although the ranges may overlap. Reports of B cell numbers were more variable for BLV-infected cattle without PL. In at least two reports, B cell numbers were reported to be decreased in BLV+, non-PL cattle (Meirom et al, 1993, Taylor et al., 1992). Variations in sample size, duration of infection, natural vs. experimental infection, failure to control for the age-related decline in total lymphocyte counts may all have contributed to these differences. Alternatively, these reported differences may reflect biological reality~i.e, that some BLV+, non-PL cattle have an Increase in B cell numbers, while others have a decrease. This possibility was alluded to by Meirom et al. (1993), who suggested that genetic differences may cause B cell number to decrease in some animals. One straightfonward way that such genetic differences might be manifested would be In an enhanced ability to present certain BLV antigens on MHC Class I molecules, which could render BLV- infected cells more prone to destruction in these animals. It would therefore be of interest to detemriine whether BLV-infected cattle with increased B cell counts have relatively weak cytotoxic T cell reactivity to BLV compared to hematologically normal animals.
B cell phenotypes In addition to being increased in number, B cells from BLV-infected cattle can have altered morphology and surface molecule expression. Possum et al. (1988) found that 3.4 to 9.7% of B cells from BLV-infected cattle with PL were enlarged blastoid type cells, compared to 0.3% for a single control animal. B cells in freshly isolated PBMC from lymphocytotic cattle also have increased levels of surface Ig (Possum et al., 1988, Matheise et al., 1992) and MHC class II molecules 37
(Fossum et al., 1988, Stone et al., 1995). In addition, a significant increase in the percentage of B cells expressing MHCII was found in freshly isolated PBMC of BLV+, non-PL cattle compared to uninfected controls (Isaacson et al., in preparation). Increased expression of MHCII molecules is an early manifestation of the B cell activation process, and cross-linking of class II molecules on resting human B cells is associated with phosphatidyl inositol turnover, protein tyrosine phosphorylation, and proliferation (Lane et al., 1990). Increased numbers of mitogen-stimulated B cells from BLV-infected cattle with PL were also found to express the inducible alpha chain of the IL-2 receptor (CD25), suggesting that BLV infection rendered B cells more sensitive to activation by T cells (Stone et al., 1994). A similar increase in B cell CD25 expression was also observed in cultured PBMC from BLV+, non-PL cattle, suggesting that an enhanced expression of molecules important in T cell-B cell interactions may not be sufficient for the development of PL (Isaacson et al., in preparation). The development of PL in BLV-infected cattle and sheep is also characterized by a dramatic expansion of the CD5+ B cell population (Letesson et al., 1990a, 1990b, Matheise et al., 1992). The CDS marker is nomnally expressed on all mature T cells and on a small proportion of B cells (Ly1+ B cells in mice) (Hayakawa and Hardy, 1988). A minor subpopulation of CD5+ B cells is normally found in newborn mice or human fetus and is believed to derive from a distinct cell lineage. Increased numbers of CD5+ B cells are present in human B cell chronic lymphocytic leukemia and autoimmune diseases such as Sjogren's syndrome, systemic lupus erythematosus and rheumatoid arthritis, as well as in some HIV- infected individuals (Raveche, 1990). The function of CDS on B cells is unclear, but there is evidence that it is a marker of B cell activation, since CDS can be induced on CDS-negative B cells by 38 phorbol ester or other stimulatory conditions (Wemer-Fabre 1989). A monoclonal antibody directed against bovine CDS enhanced mitogen-induced blastogenesis and IL-2 production in vitro, but did not directly stimulate cell division (Letesson et al, 1990a). Interestingly, an increased percentage (>15%) of CD5+ B cells in lymphocytotic cattle were shown to be activated cells with respect to Increased DNA content and responsiveness to an IL-2-containing supernatant (Maltheise et al., 1992). However, It is currently unknown whether BLV preferentially infects CD5+ B cells, resulting in their proliferation in some animals, or whether BLV- infected B cells express CDS after becoming infected. A third possibility is that CD5+ B cells are involved in Immune responses against BLV antigens, as has been suggested in other retroviral infections (Raveche, 1990). Since CDS- B cells were also increased in lymphocytotic cattle (Meirom et al., 1993), and BLV provirus has been detected by PCR in both CD5+ and CDS- B cells (Schwartz et al., 1994a), a restricted BLV tropism for CDS+ B cells is an unlikely explanation for the BLV- induced CD5+ B cell expansion. The ligand for CDS on T cells Is the CD72 protein found exclusively on B cells (Van de Velde et al., 1991), raising the intriguing possibility that enhanced CDS expression may tend to increase the activation state of BLV-infected B cells via autostimulatory interactions between CDS and CD72 on the same cell (Maltheise et al., 1992).
T cell numbers Compared to B cell numbers, there are more discrepancies in these reports of T cell numbers in BLV-infected cattle. One area of agreement is that the percentage of circulating T cells Is decreased in BLV-infected cattle with PL. However, since B cell numbers are often dramatically increased in these animals, T cell percentages would be expected to decrease, even though absolute T cell 39 numbers may be increased. As shown in Table 3, absolute T cell numbers have been reported to be decreased (Gatei et al.,1989), unchanged (Taylor et al., 1992), or Increased (Stone et a!., 1994,1995) in BLV-infected cattle with PL. As mentioned for B cell data, these variations may be due to biological variation, or may reveal flaws in experimental design. For example, in one study where absolute T cell numbers were reported to be decreased in lymphocytotic cattle (Gatei et al., 1989), the controls were much younger animals that would have been expected to have higher lymphocyte numbers because of the age-related decrease in lymphocyte counts that normally occurs in cattle. Ratios of GD4/CD8 T cells appear to be normal in BLV-infected non-PL cattle (Stone et al., 1994, 1995), but B cell/T cell ratios increased from less than 1:1 to as high as 5:1 in animals with PL
Table 3. Reported T cell percentages and/or absolute numbers in control vs. BLV- infected cattle with or without PL. See Table 2 for legend.
Control(n) BLV+, non-PL(n) BLV+, PL(n) Reference
13-26% (6) ND 4-27% (8) Takashima et al., 1977
31% (5) 29% (11) 14% (10) Esteban et al., 1985
63% (20) ND 35% (3) Paul et al., 1979
60-68% (12) ND 3x decrease (4) Possum et al., 1988
56%(4) 66%' (4) 34%* (3) Williams et al., 1988
3014 (6) 2196 (6) 1921 (6) Gatei et al., 1989
51%/2685 (15) 56% 72081* (16) 19%/2931 (11) Taylor et al., 1992
2250 (7) 1640 (7)* 1100 (2)* Sordillo et al., 1994
2170 (6) ND 4565 (8)* Stone et al., 1994
1835 (6) ND 2800 (8)* Stone et al., 1995 40
(Kumar et al., 1978, Gatei et al., 1989, Stone et al., 1994). The effect of such an alteration in normal lymphocyte proportions on T cell-B cell Interactions In vivo is unknown.
Effects on innate immunity There have been relatively few studies of the effect of BLV infection on cells of the Innate immune system. The natural cytotoxic activity of PBMC against BLV- infected FLK cells was significantly reduced in BLV-infected cattle with tumors or PL compared to non-PL or uninfected cattle (Yamamoto et al., 1984). Most sera from cattle in the terminal stages of BLV lymphoma inhibited the phagocytic activity of nomial bovine neutrophils (Takamatsu et al., 1988), but clinically healthy cows seropositive for BLV had increased in vitro neutrophil migration and production of superoxide anion, suggesting neutrophil activation (Kehrii et al., 1991). Similar findings were reported In a group of BLV-inoculated cattle, which had a transient increase in random neutrophil migration during the first year p.i. (Flaming et al., in press). BLV infection has also been shown to alter cytokine responses of bovine peripheral blood monocytes to LPS stimulation in vitro. Increased amounts of IL-1 beta and TNF were detected in supematants of LPS-treated monocytes isolated from BLV-infected cattle compared to those of uninfected animals (Weriing et al., 1995), corresponding with a previous report that monocytes may be a reservoir for BLV infection (Heeney et al., 1992). Both IL-1 and TNF can stimulate T and B cell proliferation (Ranges et al., 1988), which might contribute to the BLV-induced increase in these cell populations seen in some animals. Most recently, freshly isolated monocytes/macrophages from BLV-infected cattle with tumors or PL were shown to contain increased interieukin 10 mRNA compared to non-PL or uninfected animals (Pyeon et al., 1996). IL-10 is known to decrease the production 41
of cytokines such as IL-2 and gamma interferon gamma by T helper 1 (TH1) cells in mice, and levels of mRNA coding for these two cytokines were decreased in PBMC from BLV-lnfected cattle with PL These observations support a role for monocytes/macrophages in the progression of BLV-induced disease, and suggest that TH1-type responses may be inhibited In BLV infected cattle with PL or tumors.
Effects on acquired immunity Many studies of the alterations in lymphocyte subpopulations that occur in BLV-lnfected cattle have been purely descriptive, without consideration of the functional implications of phenotypic changes. However, the central role played by B cells in BLV infection implies that B cell functions may be affected. The most obvious abnormalities in B cell function are seen in cattle suffering from the temninal stages of BLV-induced lymphoid tumors. Serum IgM was found to be decreased or absent in 80% of cows with lymphoma (Trainin et al., 1976), and no IgM-producing cells were found in the tumorous lymph nodes of BLV-infected cattle (Trainin et al., 1971). Cows with lymphoma also had impaired primary IgM responses to a synthetic peptide antigen and to human serum albumin (Trainin 1976). There was also some evidence that antibodies of the IgM class had an abnomial structure based on altered electrophoretic mobility in cattle with lymphoma (Ungar-Waron et al., 1978). Cell-mediated immune functions are also clearly suppressed in cattle with BLV-induced tumors. Eight of thirteen BLV-infected cattle In the terminal stages of lymphoma responded pooriy to intradermal injection of mitogen or extracts from tumors or nonnal lymphoid cells (Jacobs et al., 1981). The mechanlsm(s) behind this apparent immunosuppression remain unclear; however, suppression due to soluble serum factors is commonly seen in the serum of animals and humans with 42
advanced cancer. A soluble inhibitor of nomnai lymphocyte blastogenesis has been detected in sera from BLV-infected cattle with lymphoma (Jacobs et al., 1981); however, sera from most cattle with non-BLV sporadic lymphoma also suppressed blastogenesis, suggesting that the factor was not BLV-specific (Takamatsu et al., 1988). Therefore, although immunosuppression is commonly seen in EBL, it is not necessarily an Integral feature of BLV Infection per se. Although there is no doubt that cattle in the terminal stages of BLV-induced lymphoma are immunosuppressed, the data are much less clear in BLV-infected cattle that remain clinically healthy. Similar to BLV-infected cattle with tumors, those with PL also have lower levels of semm IgM, but serum IgG levels have been reported to be unaffected (Meirom et al., 1985, Heeney et al., 1988, Gatei et al., 1990) or increased (Atluru et al., 1979). These findings have been interpreted as evidence that BLV preferentially inhibits IgM-producing cells. Another possibility that should be considered, however, is that BLV infection may stimulate B cell differentiation and antibody class switching, resulting in decreased numbers of B cells expressing IgM. It has recently been shown that secondary antibody responses to recall antigens were significantly increased in a group of BLV- infected cattle compared to uninfected controls (Isaacson et al., In preparation). This adds support to the idea that instead of causing immunosuppression, BLV may cause a generalized B cell activation under some conditions. The limited data on cell-mediated responses of clinically healthy, BLV- infected cattle appear to be in conflict. There have been several reports that lymphocytes from BLV-infected cattle with PL have reduced blastogenic responses to mitogens, as measured by stimulation indices (Weiland and Straub, 1976, Takashima et al., 1980, Villouta et al.,1988). These results appear to have been due to the high background levels of proliferation in BLV-infected cattle (see SLP), 43 since lymphocyte proliferative responses to Con A, PHA, and PWM of BLV+, non- PL animals was comparable to that of non-infected animals when DCPM were compared (Williams et al., 1988b). A highly significant correlation was reported between BLV infection and lack of recovery from Trichophyton verrucosum, a chronic fungal infection (Brenner et al., 1989), implying that cell-mediated immune functions may have been compromised. However, an early study found a significant enhancement of delayed type hypersensitivity responses in BLV- infected cattle following immunization with Mycobacterium microti (Ressang et al., 1980). More recently, an increased percentage of T cells have been found to express MHCII molecules in cultured PBMC from BLV-infected cattle, whether they have PL (Stone et al., 1995) or not (Isaacson et al., in preparation). All though the function of MHCII on T cells is unknown, it is though to be indicative of T cell activation (Odum et al., 1993). Sordillo et al. (1994) reported that supematants of cultured PBMC from BLV+, PL cattle contained significantly increased amounts of IL-2 compared to those of BLV+, non-PL or uninfected cattle. The results of Pyeon et al. diverged somewhat from this, showing that greatly increased levels of IL-2R and IFNg mRNA were present in uncultured PBMC of BLV+, non-PL animals compared to BLV+, PL cattle or uninfected controls. Since the latter study did not measure IL-2 production, it is unclear whether the different results obtained for BLV+, non-PL cattle were due to differences in the rate of translation of the IL-2 mRNA, or perhaps to biological variation. However, these studies provide further evidence that BLV infection can induce abnormalities in the acquired immune system of cattle in the absence of tumors. 44
Summary
BLV is a member of a group of retroviruses that occasionally cause tumors, but do not contain a typical retroviral oncogene or integrate at preferred sites in the host genome. Thus these viruses may have unique mechanisms of tumorigenesis.
With other oncoviruses infections such as feline leukemia virus, the host immune system is often adversely affected, even in the absence of neoplastic disease. In contrast, while there is clear evidence of immunosuppression in BLV-infected cattle with tumors, it has been more difficult to demonstrate immunosuppression in clinically healthy, BLV-infected cattle. Instead, cin accumulating body of data suggests that BLV-infected cattle without tumors may have a generalized immune activation, but the implications of these findings on bovine health are unknown.
Because of the continuing economic importance of BLV infection to the U.S. cattle industry, as well as the value of BLV infection as a model of HTLV infection in humans, it will be important to continue investigations of the effect of BLV on host immune functions.
Conclusions
Both BIV and BLV infect cells of the circulating mononuclear cell population
In cattle. Based on comparisons to other retroviruses, it would be expected that both might have immunosuppressive effects on the host immune system. In cattle infected with BIV for several months to several years, some depression of neutrophil and lymphocyte function has been observed. However, very little is known about the effects of BIV infection on acquired Immune responses. In the 45 case of BLV, only cattle in the terminal stages of cancer are obviously immunosuppressed, whereas the evidence for depressed immune function in clinically healthy, BLV-infected cattle is being outweighed by studies showing a potential enhancement of various immune functions. Thus, the literature on these two viruses to date shows that BIV has a relatively mild suppressive effect on some host Immune functions, while BLV appears to enhance many immune functions in the majority of infected animals.
References
Adam, E., Kerkhofs, P., Mammericks, M., Kettmann, R., Bumy, A., Droogmans, L. and Willems, L, 1994. Involvement of the cyclic AMP-responsive element binding protein in bovine leukemia virus expression in vivo. J. Virol. 68:5845-5853.
Alexandersen, S., Carpenter, S., Christensen, J., Storgaard, T., Viuff, B., Wannemuehler, Y., Belousov, J. and Roth, J.A., 1993. Identification of alternatively spliced mRNAs encoding potential new regulatory proteins in cattle infected with bovine leukemia virus. J. Virol. 67:39-52.
Amborski, G.F., Lo, J.L. and Seger, C.L, 1989. Serological detection of multiple retroviral infections in cattle: Bovine leukemia virus, bovine syncytial virus and bovine visna virus. Vet. Micro. 20:247-253.
Atkinson, B., Liu, Z.Q. and Wood, C., 1992. Use of bacterial trpE fusion vectors to express and characterize the bovine immunodeficiency-like virus core protein. J. Virol. Methods 36:35-49.
Atluru, D., Johnson, D.W., Paul, P.S. and Muscoplat, C.C., 1979. B lymphocyte differentiation, using pokeweed mitogen stimulation: In vitro studies in leukemic and nomnal cattle. Am. J. Vet. Res. 40:515-520.
Baliga, V. and Ferrer, J.F., 1977. Expression of the bovine leukemia virus and its internal antigen in blood lymphocytes. Proc. Soc. Exp. Biol. Med. 156:388-391.
Battles, J.K., Hu, M.Y., Rasmussen, L., Tobin, G.J. and Gonda, M.A., 1992. Immunological characterization of the gag gene products of bovine immunodeficiency virus. J. Virol. 66:6868-6877. 46
Ban, J., Portetelle, D., Altaner, C., Horion, B., Milan, D., Krchnak, V., Bumy, A. and Kettmann, R., 1993. Isolation and characterization of a 2.3-kilobase-pair cDNA fragment encoding the binding domain of the bovine leukemia virus receptor. J. Virol. 67:1050-1057.
Bendixen, H.J., 1965. Bovine enzootic leukosis. Advances in Veterinary Science and Comparative Medicine 10:129-204.
Bishop, S.A., Williams, N.A., GruffyDD-Jones, T.J., Harbour, D.A. and Stokes, G.R., 1992. An eariy defect in primary and secondary T cell response in asymptomatic cats during acute feline immunodeficiency virus (FIV) infection. Clin. Exp. Immunol. 90:491-496.
Black, J., 1989. Report of the committee on bluetongue and bovine retrovirus. 93rd Annual Meeting of the United States Animal Health Assoc. 145-152.
Braun, M.J., Lahn, S., Boyd, A.L, Kost, T.A., Nagashima, K. and Gonda, M.A., 1988. Molecular cloning of biologically acive proviruses of bovine immunodeficiency-like virus. Virology 167:515-523.
Brenner, J., Van-Haam, M., Savir, D. and Trainin, Z., 1989. The implication of BLV infection in the productivity, reproductive capacity and survival rate of a dairy cow. Vet. Immunol. Immunopathol. 22:299-305.
Brooks, P.A., Nyborg, J.K. and Cockerell, G.L, 1995. Identification of an NF-kB binding site in the bovine leukemia virus promoter. J. Virol. 69:6005-6009.
Brownlie, J., Collins, M.E. and Heaton, P., 1994. Bovine immunodeficiency-like virus~a potential cause of disease in cattle? Vet. Record 134:289-291.
Bruck, C., Mathot, S., Portetelle, D., Berte, C., Franssen, J.D., Herion, P. and Bumy, A., 1982a. Monoclonal antibodies define eight independent antigenic regions on the bovine leukemia virus (BLV) envelope glycoprotein gp51. Virology 122:343- 352.
Bruck, C., Portetelle, D., Bumy, A. and Zavada, J., 1982b. Topographical analysis by monoclonal antibodies of BLV gp51 epitopes involved in viral functions. Virology 122:353-362.
Bumy, A, Bruck, C., Chantrenne, H., Cleuter, Y., Dekegel, D., Ghysdael, J., Kettmann, R., Leclercq, M., Leunen, J., Mammerickx, M. and Portetelle, D., 1980. Bovine leukemia virus: Biology and epidemiology. In Klein, G. (ed.) Viral Oncology. New York, Raven Press, 231-289.
Bumy, A., Cleuter, Y., Kettmann, R., Mammerickx, M., Marbaix, G., Portetelle, Van Den Broeke, A., Willems, L. and Thomas, R., 1988. Bovine leukaemia: facts and hypotheses derived from the study of an infectious cancer. Vet. Micro. 17:197-218. 47
Cann, A J. and Chen, I.S.Y., 1990. Human T-cell leukemia virus types I and II. In: Virology, 2nd ed. Raven Press, New York, pp. 1501-1527.
Carpenter, S. and Alexanderson, S., 1992. Pathogenesis of equine Infectious anemia virus infection. Seminars In Virol. 3:157-166.
Carpenter, S., Miller, L.D., Alexandersen, S., Whetstone, C.A., Van Der Maaten, M.J., Viuff, B., Wannemuehler, Y., Miller, J.M. and Roth, J.A., 1992. Characterization of early pathogenic effects after experimental infection of calves with bovine immunodeflclency-like virus. J. Virol. 66:1074-1083.
Carpenter, S., Nadln-Davls, S.A., Wannemuehler, Y. and Roth, J.A., 1993. Identification of transactivation-response sequences In the long terminal repeat of bovine Immunodeficiency-like virus. J. Virol. 67:4399-4403.
Chadwick, B.J., Coelen, R.J., Wilcox, G.E. and Kertayadnya, G., 1995. Nucleotide sequence analysis of Jembrana disease virus: a bovine lentivlrus associated with an acute disease syndrome. J. Gen. Virol. 76:1637-1650.
Chandler, S. and Gilman, J.P.W., 1974. Relationship between lymphocytosis and DMA synthetic lymphocytes. Can. J. Comp. Med. 38:430-436.
Chatterjee, R., Gupta, P., Kashmiri, S.V.S. and Ferrer, J.F., 1985. Phytohemmagglutlnin activation of the transcription of the bovine leukemia virus genome requires de novo protein synthesis. J. Virol. 54:860-863.
Chen, P., Liu, Z.Q. and Wood, C., 1994. Use of TrpE fusion protein to identify antigenic domains within the BIV envelope protein. J. Virol. Methods 47:331-343.
Chen, L. and Frankel, A.D., 1994. An RNA-blndIng peptide from bovine immunodeficiency virus Tat protein recognizes an unusual RNA structure. Biochemistry 33:2708-2715.
Cockerell, G.L. and J. Rovnak, J., 1988. The correlation between the direct and indirect detection of bovine leukemia virus infection in cattle. Leukemia Res. 12:465-469.
Cockerell, G.L., Jensen, W.A., Rovnak, J., Ennis, W.H. and Gonda, M.A., 1992. Seroprevalence of bovine immunodeficiency-like virus and bovine leukemia virus in a dairy herd. Vet. Micro. 31:109-116.
Coffin, J.M., 1992. Structure and classification of Retroviruses. In: J.A. Levy, (Editor). The Retroviridae, Volume 1, Plenum Press, New York, pp. 19-49.
Cotter, F.E., 1993. Molecular pathology of lymphomas. Cancer Surv. 16:157-174.
Coulston, J., Naif, H., Brandon, R., Kumar, S., Khan, S., Daniel, R.C.W. and Lavin, M.F., 1990. Molecular cloning and sequencing of an Australian isolate of proviral 48
bovine leukaemia virus DNA: comparison with other Isolates. J. Gen. Virol. 71:1737-1746.
Cullen, B.R., 1992. Mechanism of action of regulatory proteins encoded by complex retroviruses. Microbiol. Reviews 56:375-394.
Da., Y., Shanks, R.D., Stewart, J.A. and H.A. Lewin, H.A., 1993. Milk and fat yields decline In bovine leukemia virus-infected Holstein cattle with persistent lymphocytosis. Proc. Natl. Acad. Sci. 90:6538-6541.
Derse, D., 1987. Bovine leukemia virus transcription Is controlled by a virus- encoded trans-acting factor and by cis-acting response elements. J. Virol. 61:2462- 2471.
Derse, D., 1988. Trans-acting regulation of bovine leukemia virus mRNA processing. J. Virol. 62:1115-1119.
Deshayes, L, Levy, D., ParodI, A.L. and Levy, J.P., 1980. Spontaneous immune response of bovine leukemia virus-Infected cattle against five different viral proteins. Int. J. Cancer 25:503-508.
Detllleux, J.C., Kehrii, M.E., Freeman, A.E., Whetstone, C.A. and Kelley, D.H., 1995. Two retroviral infections of periparturient Holstein cattle: A phenotypic and genetic study. J. Dairy Sci. 78:2294-2298.
Djilall, S., ParodI, A. and Levy, D., 1987. Bovine leukemia virus replicates In sheep B lymphocytes under a T cell released factor. Eur. J. Cancer Clin. Oncol. 23:81-85.
Driscoll, D.M., Baumgartener, L.E. and Olsen, C., 1977. Concanavalln A and the production of bovine leukemia virus antigen in short-term lymphocyte cultures. J. Natl. Cancer. Inst. 58:1513-1514.
Esteban, E.N., Thom, R.M. and J.F. Ferrer, J.F., 1985. Characterization of the blood lymphocyte population in cattle Infected with the bovine leukemia virus. Cancer Res. 45:3225-3230.
Fen-er, J.F., Bhatt, D.M., Abt, D.A., Marshak, R.R. and Baliga, V.L., 1975. Serological diagnosis of infection with the putatuve bovine leukemia virus. Cornell Vet. 65:527- 542.
Fen-er, J.F. and Diglio, C.A. 1976. Development of an In vitro infectivity assay for the C-type bovine leukemia virus. Cancer Res. 36:1068-1073.
Ferrer J.F., Piper, C.E., Abt, D.A. and Marshak, R.R., 1977. Diagnosis of bovine leukemia virus infection: Evaluation of serologic and hematologic tests by a direct infectivity detection assay. Am. J. Vet. Res. 38:1977-1981. 49
Ferrer, J.F., Marshak, R.R., Abt, D.A. and Kenyon, S.J., 1979. Relationship between lymphosarcoma and persistent lymphocytosis In cattle: a review. J. Amer. Vet. Med. Assoc. 175:705-708.
Ferrer, J.F., 1980. Bovine lymphosarcoma. Adv. Vet. Sci. Comp. Med. 24:1-68.
Flaming, K., Van Der Maaten, M., Whetstone, C., Carpenter, S., Frank, D. and Roth, J., 1993. The effect of bovine immunodeficiency-like virus Infection on immune function in experimentally infected cattle. Vet. Immunol. Immunopathol. 36:91-105.
Flaming, K.P., Frank, D.E., Carpenter, S. and Roth, J.A. Longitudinal studies of immune function in cattle experimentally infected with bovine immunodeficiency like virus and/or bovine leukemia virus. Vet. Immunol. Immunopathol. In press.
Fong, S.E., Pallansch, LA., Mikovits, J.A., Lackman-Smith, C.S., Ruscetti, F.W. and Gonda, M.A., 1995. Cis-acting regulatory elements in the bovine immunodeficiency-like virus long terminal repeat. Virology 209:604-614.
Fossum, C., Bumy, A., Portetelle, D., Mammerickx, M. and Morein, B., 1988. Detection of B and T cells, with lectins or antibodies, in healthy and bovine leukemia virus-infected cattle. Vet Immunol. Immunopathol. 18:269-278.
Gatei, M.H., Brandon, R.B., Naif, H.M., McLennan, M.W., Daniel, R.C.W. and Lavin, M.F., 1989. Changes In B cell and T cell subsets in bovine leukaemia virus-infected cattle. Vet. Immunol. Immunopathol. 23:139-147.
Gatei, M.H., Lavin, M.F. and Daniel, R.C.W., 1990. Serum immunoglobulin concentrations in cattle naturally infected with bovine leukemia virus. J. Vet. Med. B 37:575-580.
Gatei, M.H., Naif, H.M., Kumar, S., Boyle, D.B., Daniel, R.C.W., Good, M.F. and Lavin, M.F., 1993. Protection of sheep against bovine leukemia vims (BLV) infection by vaccination with recombinant vaccinia viruses expressing BLV envelope glycoproteins: correlation of protection with CD4 T-cell response to gp51 peptide 51-70. J. Virol. 67:1803-1810.
Garvey, K.J., Oberste, M.S., Elser, J.E., Braun, M.J. and Gonda, M.A., 1990. Nucleotide sequence and genome organization of biologically active proviruses of the bovine immunodeficiency-like vims. Virology 175:391-409.
Gonda, M.T., Braun, M.J., Carter, S.G., Kost, T.A., Bess, J.W., Arthur, L.O. and Van Der Maaten, M.J., 1987. Characterization and molecular cloning of a bovine lentivirus related to human immunodeficiency vims. Nature 330:388-391.
Gupta, P. and Ferrer, J.F., 1982. Expression of bovine leukemia vims genome is blocked by a nonimmunoglobulin protein in plasma from infected cattle. Science 215:405-407. 50
Gupta, P., Kashmiri, S.V.S. and Ferrer, J.F., 1984. Transcriptional control of the bovine leukemia virus genome: Role and characterization of a non- immunoglobulin plasma protein from bovine leukemia virus-infected cattle. J. Virol. 50:267-270.
Gupta, P., Chattetjee, R. and Cai, Q., 1989. Prevalence of the plasma bovine leukemia virus blocking factor in cattle from a commercial dairy herd. Vet. Record. 125:5-6.
Hayakawa, K, and Hardy, R.R., 1988. Normal, autoimmune, and malignant CD5+ B cells: the Ly-1 B lineage? Ann. Rev. Immunol. 6:197-218.
Heeney, J.L, Valli, V.E. and Montesanti, J., 1988. Alterations in humoral immune response to bovine leukaemia virus antigens in cattle with lymphoma. J. Gen. Virol. 69:659-666.
Heeney, J.L. and Valli, V.E.O., 1990. Transformed phenotype of enzootic bovine lymphoma reflects differentiation-linked leukemogenesis. Lab. Invest. 62:339-346.
Heeney, J.L., Valli, P.J.S., Jacobs, R.M. and Valli, V.E.O., 1992. Evidence for bovine leukemia virus infection of peripheral blood monocytes and limited antigen expression in bovine lymphoid tissue. Lab. Invest. 66:608-617.
Huber, N.L., DiGiacomo, R.F., Evemiann, J.F. and Studer, E., 1981a. Bovine leukemia virus infection In a large Holstein herd: Cohort analysis of the prevalence of antibody-positive cows. Am. J. Vet. Res. 42:1474-1476.
Huber, N.L., DiGiacomo, R.F., Evenmann, J.F. and Studer, E., 1981b. Bovine leukemia virus infection in a large Holstein herd: prospective comparison of production and reproductive performance in antibody-negative and antibody- positive cows. Am. J. Vet. Res. 42:1477-1481.
Jacobs, R.M., 1980. Bovine lymphoma. In: R.G. Olsen, S. Krakowka, and J.R. Blakeslee (Editors.), Comparative Pathobiology of Viral Diseases, CRC Press, Inc., Boca Raton, Florida, pp. 22-51.
Jacobs, R.M., Valli, V.E.O. and Wilkie, B.N., 1981. Response of cows with lymphoma to the intradermal injection of tumor cell antigens and phytohemagglutinin. Can. J. Comp. Med. 45:43-50.
Jacobs, R.M, Pollari, F.L., McNab, W.B. and Jefferson, B., 1995. A serological survey of bovine syncytial virus In Ontario: association with bovine leukemia and immunodeficiency-like viruses, production records, and management practices. Can. J. Vet. Res. 59:271-278.
Jensen, W.A., Sheehy, S.E., Fox, M.H., Davis, W.C. and Cockerell, G.L., 1990. In vitro expression of bovine leukemia virus in isolated B-lymphocytes of cattle and sheep. Vet. Immunol. Immunopathol. 26:333-342. 51
Jensen, W.A., Rovnak, J. and Cockerell, G.L, 1991. In vivo transcription of the bovine leukemia virus tax/rex region in nonnal and neoplastic lymphocytes of cattle and sheep. J. Virol. 65:2484-2490.
Jerabek, L., Gupta, P., and Ferrer, J.F., 1979. An infectivity assay for bovine leukemia virus using the immunoperoxidase technique. Cancer Res. 39:3952- 3954.
Johnson, R. and Kaneene, J.B., 1991. Bovine leukemia virus. Part 1. Descriptive epidemiology, clinical manifestations, and diagnostic tests. Compend. Cont. Educ. Pract. Vet. 13:315-326.
Kaneene, J.B., Nicol, J. and Gibson, C.D., 1983. Prevalence of bovine leukemia virus infection in Michigan dairy cattle. Michigan State University Extension Bulletin No. E-1674, pp. 1-2.
Kehrii, M.E., Weigel, K.A., Freeman, A.E., Thurston, J.R. and Kelley, D.H., 1991. Bovine sire effects on daughters' in vitro blood neutrophil functions, lymphocyte blastogenesis, serum complement, and conglutinin levels. Vet, Immunol. Immunopathol. 27:303-319.
Kenyon, S.J., and Piper, C.E., 1977a. Cellular basis of persistent lymphocytosis In cattle infected with bovine leukemia virus. Infect. Immun. 16:891-897.
Kenyon, S.J., and Piper, C.E., 1977b. Properties of density gradient-fractionated peripheral blood leukocytes from cattle infected with bovine leukema virus. Infect. Immun. 16:898-903.
Kettmann, R., Portetelle, D., Mammerickx, M., Cleuter, Y., Dekegel, D., Galoux, M., Ghysdael, J., Bumy, A. and Chantrenne, H., 1976. Bovine leukemia virus: an exogenous RNA oncogenic virus. Proc. Natl. Acad. Sci. 73:1014-1018.
Kettmann, R., Cleuter, Y., Mammerickx, M., Meunier-Rotival, M., Bemardi, G., Bumy, A. and Chantrenne, H., 1980. Genomic integration of bovine leukemia virus provirus: comparison of persistent lymphocytosis with lymph node tumor fomn of enzootic bovine leukosis. Proc. Natl. Acad. Sci. USA 77:2577-2581.
Kettmann, R, Cleuter, Y., Gregoire, D. and Bumy, A., 1985. Role of the 3' open reading frame region of bovine leukemia virus in the maintenance of cell transformation. J. Virol. 54:899-901.
Kumar, S.P., Paul, P.S., Pomeroy, K.A., Johnson, D.W., Muscoplat, C.C., Van Der Maaten, M.J., Miller, J.M. and Sorenson, D.K., 1978. Frequency of lymphocytes bearing Fc receptors and surface membrane immunoglobulins in nomnal, persistently lymphocytotic and leukemic cows. Amer. J. Vet. Res.39:45-49. 52
Lagarias, D.M. and Radke, K., 1989. Transcriptional activation of bovine leukemia virus in blood cells from experimentally infected asymptomatic sheep with latent infections. J. Virol. 63:2099-2107.
Lane, P.J.L., McConnell, F.M., Schieven, G.L., Clark, E.A. and Ledbetter, J.A., 1998. The role of class II molecules in human B cell activation. Association with phosphatidyl inositol turnover, protein tyrosine phosphorylation, and proliferation. J. Immunol. 144:3684-3692.
Lettesson, J.J., Mager, A., Didembourg, C. and Depelchin, A., 1990a. Monoclonal antigody specific for bovine CDS antigen which enhances mitogen-induced blastogenesis and IL-2 production. Vet. Immunol. Immunopathol. 25:249-257.
Lettesson, J.J., Mager, A., Mammerickx, M., Bumy, A. and Delpinchen, A., 1990b. B cells from bovine leukemia virus (BLV) infected sheep with hematological disorders express the CD5 T cell mariner. Leukemia 4:377-379.
Lewin, H.A. and Bemoco, D., 1986. Evidence for BoLA-iinked resistance and susceptibility to subclinical progression of bovine leukaemia virus infection. Animal Genetics 17:197-207.
Lewin, H.A., Wu, M.C., Stewart, J.A and Nolan, T.J., 1988. Association between BoLA and subclinical bovine leukemia vims infection in a herd of Holstein-Friesian cows. Immunogenetlcs 27:338-344.
Lewin, H.A., 1994. Host genetic mechanism of resistance and susceptibility to a bovine retroviral infection. Animal Biotechnology 5:183-191.
Malmquist, W.A., Van Der Maaten, M.J. and Boothe, A.D., 1969. Isolation, Immunodiffusion, Immunofluorescence, and electron microscopy of a syncytial virus of lymphosarcomatous and apparently normal cattle. Cancer Res. 29:188- 200.
Mathelse, J.P., Delcommenne, M., Mager, A., Didembourg, C.H., and Letesson, J.J., 1992. CD5+ B cells from bovine leukemia virus infected cows are activated cycling cells responsive to interieukin 2. Leukemia 6:304-309.
Mansky, L.M. and Temin, H.M., 1994. Lower mutation rate of bovine leukemia virus relative to that of spleen necrosis virus. J. Virol. 68:494-499.
Mariott, S.J., Lindholm, P.F., Reid, R.L. and Brady, J.N., 1991. Soluble HTLV-I Taxi protein stimulates proliferation of human peripheral blood lymphocytes. New. Biol. 3:678-686.
Mariott, S.J., Trinh, D. and Brady, J.N., 1992. Activation of interieukin-2 receptor alpha expression by extracellular HTLV-I Taxi protein: a potential role in HTLV-I pathogenesis. Oncogene 7:1749-1755. 53
Martin, SJ., O'Neill, T.P., Bilello, J.A. and Eiseman, J.L, 1991. Lymphocyte transfomiation abnomnallties in bovine immunodeficiency-like virus infected calves. Immunol. Let. 27:81-84.
Matheise, J.P., Delcommenne, M., Mager, A., Didembourg, C.H. and J.J. Letesson, J.J., 1992. CD5+ B cells from bovine leukemia virus infected cows are activated cycling cells responsive to interieukin 2. Leukemia 6:304-309.
Mauxion, F., Jamieson, C., Yoshida, M., Arai, K.I. and Sen, R., 1991. Comparison of constitutive and inducible transcriptional enhancement mediated by kB-related sequences: modulation of activity in B cells by human T-cell leukemia virus type I tax gene. Proc. Natl. Acad. Sci. 88:2141-2145.
McDonald, H.C. and Ferrer, J.F., 1976. Detection, quantitation, and characterization of the major intemal virion antigen of the bovine leukemia virus by radioimmunoassay. J. Natl. Cancer. Inst. 57:875-882.
McNab, W.B., Jacobs, R.M. and Smith, H.E., 1994. A serological survey for bovine immunodeficiency-like virus in Ontario dairy cattle and associations between test results, production records and management practices. Can. J. Vet. Res. 58:36-41.
Meirom, R., Brenner, J., Gluckman, A., Avraham, R. and Trainin, Z., 1985. Humoral and cellular responses in calves experimentally infected with bovine leukemia virus (BLV). Vet. Immunol. Immunopathol. 9:105-114.
Meirom, R., Brenner, J. and Trainin, Z., 1993. BLV-infected lymphocytes exhibit two pattems of expression as determined by Ig and CD5 markers. Vet. Immunol. Immunopathol. 36:179-186.
Meyaard, L., Schuitemakler, H. and Miedema, F., 1993. T-cell dysfunctioin in HIC infection: anergy due to defective antigen-presenting cell function?
Miedema, F., Chantal Petit, A.J., Terpstra, F.G., Eeftinck Schattenkerk, J.K.M., deWolf, F., Al, B.J.M., Roos, M., Lange, J.M.A., Danner, S.A., Goudsmit, J. and Schellekens, R.T.A., 1988. Immunological abnormalities in human immunodeficiency virus (HIV)-infected asymptomatic homosexual men. J. Clin. Invest. 82:1908-1914.
Miller, J.M., Miller, L.D., Olson, C. and Gillette, K.G., 1969. Virus-like particles in phytohemagglutinin-stimulated lymphocyte cultures with reference to bovine lymphosarcoma. J. Natl. Cancer Inst. 42:1297-1305.
Miller, J.M. and Van der Maaten, M.J., 1974. A complement fixation test for the bovine leukemia (C-type) virus. J. Natl. Cancer Inst. 53:1699-1702.
Miller, J.M. and Van der Maaten, M.J., 1976. Serologic detection of bovine leukemia virus infection. Vet. Micro. 1:195-202. 54
Miller, J.M., 1981. Leukosis in cattle. Calif. Vet. 8:31-33.
Miller, J.M., Schmerr, MJ.F. and Van der Maaten, M.J., 1981. Comparison of four serologic tests for the detection of antibodies to bovine leukemia virus. Amer. J. Vet. Res. 42:5-8.
Muscoplat, C.C., Johnson, D.W., Pomeroy, K.A., Olson, J.M., Larson, V.L., Stevens, J.B. and Sorenson, D.K., 1974a. Lymphocyte surface immunoglobulin: Frequency in normal and lymphocytic cattle. Am. J. Vet. Res. 35:593-595.
Muscoplat, C.C., Alhaji, I., Johnson, D.W., Pomeroy, K.A., Olson, J.M., Larson, V., Stevens, J.B. and Sorenson, D.K., 1974b. Characteristics of lymphocyte responses to phytomitogens: Comparison of responses of lymphocytes from normal and lymphocytotic cows. Am. J. Vet. Res. 35:1053-1055.
Nadin-Davis, S., Chang, S.C., Smith, H. and Jacobs, R.M., 1993. Detection of bovine immunodeficiency-like virus by the polymerase chain reaction. J. Virol. Methods 42:323-336.
Narayan, O. and Clements, J.E, 1989. Biology and pathogenesis of Antiviruses. J. Gen. Virol. 70:1617-1639.
Oberste, M.S., Greenwood, J.D. and Gonda, M.A., 1991. Analysis of the transcription pattern and mapping of the putative rev and env splice junctions of bovine immunodeficiency-like vims. J. Virol. 65:3932-3937.
Odum, N., Kanner, S.B., Ledbetter, J.A. and Svejgaard, A., 1993. MHC Class II molecules diliver costimulatory signals in human T cells throu a functional linkage with IL-2 receptors. J. Immunol. 150:5289-5298.
Ohishi, K., Suzuki, H., Yamamoto, T., Mamyama, T., Miki, K., Ikawa, Y., Numakunai, S., Okada, K., Ohshima, K. and Sugimoto, M., 1991. Protective immunity against bovine leukemia virus (BLV) induced in carrier sheep by inoculation with a vaccinia virus-BLV env recombinant: association with cell-mediated immunity. J. Gen. Virol. 72:1887-1892.
Okada, K., Ikeyama, S., Ohishi, K., Suzuki, H., Sugimoto, M., Numakunai, S., Chiba, T., Murakami, K., Davis, W.C. and Ohshima, K., 1993. Involvement of CD8+T cells in delayed-type hypersensitivity responses against bovine leukemia virus (BLV) induced in sheep vaccinated with recombinant vaccinia virus expressing BLV envelope glycoprotein. Vet. Pathol. 30:104-110.
Onuma, M., Koomoto, E., Furuyama, H., Yasutomi, Y., Taniyama, H., Iwai, H., and KawakamI, Y., 1992. Infection and dysfunction of monocytes induced by experimental inoculation of calves with bovine immunodeficiency-like virus. JAIDS 5:1009-1015. 55
Pallansch, L.A., Lackman-Smrth, C.S. and Gonda, M.A., 1992. Bovine immunodeficiency-like virus encodes factors which trans-activate the long terminal repeat. J. Virol. 66:2647-2652.
Paul, P.S., Pomeroy, K.A., Johnson, D.W., Muscoplat, C.C., Handwerger, B.S., Soper, F.F. and Sorensen, D.K., 1977. Evidence for the replication of bovine leukemia virus in the B lymphocytes. Am. J. Vet. Res. 38:873-876.
Paul, P.S., Senogles, D.R., Muscoplat, C.C. and Johnson, D.W., 1979. Enumeration of T cells, B cells, and monocytes in the peripheral blood of normal and lymphoc^otic cattle. Clin. Exp. Immunol. 35:306-316.
Perman, V., Dirks, V.A., Fangmann, G., Snyder, M.M., Sorensen, K.K., Anderson, R.K., Goltz, D.J., Larson, V.L. and Stevens, J.B., 1970. Statistical evaluation of lymphocyte values on Minesota dairy cattle. Am. J. Vet. Res. 31:1217-1222.
Pinching, A.J., 1991. Antibody responses in HIV infection. Clin. Exp. Immunol. 84:181-184.
Pollari, F.I., DiGiacomo, R.F. and Evermann, J.F., 1993. Use of survival analysis to compare cull rates between bovine leukemia virus seropositive and seronegative dairy cows. Am. J. Vet. Res. 54:1400-1403.
Pyeon, D., O'Reilly, K.L. and Splitter, G.A., 1996. Increased interleukin-10 mRNA expression in tumor-bearing or persistently lymphocytotic animals infected with bovine leukemia virus. J. Virol. 70:5706-5710.
Ranges, G., ZIotnik, A., Espevik, T., Cerami, C.A. and Palladino, M., 1988. Tumor necrosis factor alpha/cachectin is a growth factor for thymocytes. J. Exp. Med. 167:1472-1478.
Rasmussen, L, Greenwood, J.D., and Gonda, M.A., 1992. Expression of bovine immunodeficiency-like virus envelope glycoproteins by a recombinant baculovirues in insect cells. Virology 186:551-561.
Raveche, E.S., 1990. A possible immunoregulatory role for CD5+ B cells. Clin. Immunol. Immunopathol. 56:135-150.
Ressang, A.A., Rumawas, W., Rondhuis, P.R.R., Haagsma, J. and Bercovich, Z., 1980. Studies on Bovine Leukosis VIII. the serologic response of leucotic cattle to injections with Aujeszky's disease and bordetella vaccines, tetanus toxoid and porcine erythrocytes, the cutaneous reaction following administration of mycobacterium microti, and the effect of pokeweed and phytohaemagglutinin stimulation on leucotic lymphocytes. Zbl. Vet. Med. B. 27:576-588.
Rosen, C.A., Sodroski, J.G., Kettmann, R., Bumy, A. and Haseltine, W.A., 1985. Trans activation of the bovine leukemia virus long terminal repeat in BLV-infected cells. Science 227:320-322. 56
Rovid, A.H„ Carpenter, S. and Roth, J.A., 1995. Monocyte function in cattle experimentally infected with bovine immunodeficiency-like virus. Vet. Immunol. Immunopathol. 45:31-43.
Rovid, A.H., Carpenter, S., Miller, L.D., Flaming, K.P., Long, M.J., Van Der Maaten, M.J., Frank, D.E. and Roth, J.A., 1996. An atypical T-cell lymphosarcoma in a calf with bovine immunodeficiency-like virus infection. Vet. Pathol. 33:457-9.
Rovnak, J., Casey, J.W., Boyd, A.L, Gonda, M.A. and Cockerell, G.L, 1991. Isolation of bovine leukemia vims infected endothelial cells from cattle with persistent lymphocytosis. Lab. Invest. 65:192-201.
Rovnak, J., Boyd, A.L., Casey, J.W., Gonda, M.A., Jensen, W.A. and Cockerell, G.L., 1993. Pathogenicity of moleculariy cloned bovine leukemia virus. J. Virol. 67:7096- 7105.
Sagata, N., Yasunaga, T., Tsuzuku-Kawamura, J., Ohishi, K. and Ogawa, Y., 1985. Complete nucleotide sequence of the genome of bovine leukemia virus: its evolutionary relationship to other viruses. Proc. Natl. Acad. Sci. USA 82:677-681.
Schwartz, I., Bensaid, A., Polack, B., Pemn, B., Berthelemy, M. and Levy, D, 1994. In vivo leukocyte tropism of bovine leukemia vims in sheep and cattle. J. Virol. 68:4589-4596.
Schwartz, I. and Levy, D., 1994. Pathobiology of bovine leukemia virus. Vet. Res. 25:521-536.
Shearer, G.M., Bernstein, D.C., Tung, K.S.K., Via, C.S., Redfield, R., Salhuddin, S.Z. and Gallo, R.C., 1986. A model for the selective loss of major histocompatibility complex self-restricted T cell immune responses during the development of acquired immune deficiency syndrome (AIDS). J. Immunol. 137:2514-2521.
Shemnan, M.P., Ehriich, G.D., Ferrer, J.F., Sninsl^, J.J., Zandomeni, R., Dock, N.L. and Poiesz, B.J., 1992. Amplification and analysis of specific DNA and RNA sequences of bovine leukemia virus from infected cows by polymerase chain reaction. J. Clin. Microbiol. 30:185-191.
Snider, T.G., Luther, D.G., Jenny, B.F., Hoyt, P.G., Battles, J.K., Ennis, W.H., Balady, J., Blas-Machado, U., Lemarchand, T.X., and Gonda, M.A., 1996. Encephalitis, lymphoid tissue depletion and secondary diseases associated with bovine immunodeficiency virus in a dairy herd. Comp. Immunol. Microbiol. Infect. Dis. 19:117-131.
Soeharsono, S., Hartaningsih, N., Soetrisno, M., Kertayadnya, G., and Wilcox, G.E., 1990. Studies of experimental Jembrana disease in Bali cattle. I. Transmission and persistence of the infectious agent in mminants and pigs, and resistance of recovered cattle to re-infection. J. Comp. Path. 10:49-59. 57
Sordillo, L.M., Hicks, C.R., and Pighetti, G.M., 1994. Altered interleukin-2 production by lymphcyte populations from bovine leukemia virus-infected cattle. Proc. Soc. Biol. Med. 207:267-273.
Specter, S., Bendinelli, M. and Friedman, H. 1988. Nonhuman mammalian retroviruses. In: Virus-Induced Immunosuppression (S. Specter, M. Bendinelli, and H. Friedman, Editors), Plenum Press, New York and Londen, p. 395-419.
St. Cyr Coats, K., Pruett, S.B., Nash, J.W, and Cooper, C.R., 1995 Bovine immunodeficiency virus: incidence of infection In Mississippi dairy cattle. Vet. Microbiol. 42:181-189.
Stear, M.J., Dimmock, C.K., Newman, M.J. and Nicholas, F.W., 1988. BoLA antigens are associated with increased frequency of persistent lymphocytosis in bovine leukemia virus infected cattle and with increased incidence of antibodies to bovine leukemia virus. Animal Genetics 19:151-158.
Stone, D.M., McElwain, T.F. and Davis, W.C., 1994. Enhanced B-lymphocyte expression of IL-2Ra associated with T lymphocytosis in BLV-infected persistently lymphocytotic cows. Leukemia 8:1057-1061.
Stone, D., Hof, A.J. and Davis, W.C., 1995. Up-regulatlon of IL-2 receptor alpha and MHC class II expression on lymphocyte subpopulations from bovine leukemia virus infected cows. Vet. Immunol. Immunopathol. 48:65-76.
Stott, M.L, Thurmond, M.A., Theilen, G.H., Sninsky, J., Kwon, D. and Taylor, B., 1989. Infection of cattle with bovine immunodeficiency-like virus. Proc. 70th Conf. Res. Workers Anim. Dis., 238.
Stott, M.L, Thumiond, M.A., Dunn, S.J., Osbum, B.I. and Stott, J.L., 1991. Integrated bovine leukosis proviral DNA in T helper and T cytotoxic/suppressor lymphocytes. J. Gen. Virol. 72:307-315.
Suarez, D.L, Van Der Maaten, M.J., Wood, C. and Whetstone, C.A., 1993. Isolation and characterization of new wild-type isolates of bovine lentivirus. J. Virol. 67:5051- 5055.
Suarez, D.L. and Whetstone, C.A., 1995. Identification of hypervariable and consen/ed regions in the surface envelope gene in the bovine lentivirus. Virology 212:728-733.
Suarez, D.L., Van Der Maaten, M.J. and Whetstone, C.A., 1995. Improved early and long-term detection of bovine lentivirus by a nested polymerase chain reaction test in experimentally infected calves. Am. J. Vet. Res. 56:579-586. 58
Takamatsu, H., Inumaru, S. and Nakajima, H., 1988. Inhibition of in vitro imnnunocyte function by sera from cattle with bovine leukosis. Vet. Immunol. Immunopathol. 18:349-359.
Takashima, I., Olson, C. and Driscoll, D.M., 1977. B lymphocytes and T lymphocytes in three types of bovine lymphosarcoma. J. Natl. Cancer Inst. 59:1205- 1209.
Takashima, I. and Olson, C., 1980. Effect of mitogens and anti-bovine leukosis virus serums on DNA synthesis of lymphocytes from cattle. Europ. J. Cancer 16:639-645.
Takashima, I. and Olson, C., 1981. Relation of bovine leukosis virus production on cell growth cycle. Arch. Virol. 69:141-148.
Taylor, B.C., Stott, J.L, Thurmond, M.A. and Picanso, J.P., 1992. Alteration in lymphocyte subpopulations in bovine leukemia virus-infected cattle. Vet Imm. Immunopathol. 31:35-47.
Taylor, J.A. and Jacobs, R.M., 1993. Effects of plasma and serum on the in vitro expression of bovine leukemia virus. Lab. Invest. 69:340-346.
Thorn, R.M., Gupta, P., Kenyon, S.J. and Ferrer, J.F., 1981. Evidence that the spontaneous blastogenesis of lymphocytes from bovine leukemia virus-infected cattle is viral antigen specific. Infect Immun 34:84-89.
Thurmond, M.C., 1987. Economics of bovine leukemia vims infection. In: Enzootic Bovine Leukosis and Bovine Leukemia Vims, A. Bumy and M. Mammerickx (Editors), Martinus Nijhoff Publishing, Boston, pp. 71-84.
Thurmond, M.C., Carter, R.L., Picanso, J.P. and Stralka, K., 1990. Upper-normal prediction limits of lymphocyte counts for cattle not Infected with bovine leukemia vims. Am. J. Vet. Res. 51:466-470.
Tobin, G.J,, Sowder, R.C.II, Fabris, D., Hu, M.Y., Battles, J.K., Fenselau, C., Henderson, L.E. and Gonda, M.A., 1994. Amino acid sequence analysis of the proteolytic cleavage products of the bovine Immunodeficiency vims Gag precursor polypeptide. J. Virol. 68:7620-7627.
Torten, M., Franchini, M., Barlough, J.E., George, J.W., Mozes, E., Lutz, H. and Pedersen, N.C., 1991. Progressive immune dysfunction in cats experimentally infected with feline immunodeficiency vims. J. Virol. 65:2225-2230.
Trainin, Z. and Klopfer, U., 1971. Immunofluorescent studies of lymph nodes and spleens of leukotic cattle for cells producing IgM and IgG. Cancer Res. 41:1968- 1970.
TrainIn, Z., Ungar-Waron, H., Meirom, R., Bamea, A. and Sela, M., 1976. IgG and IgM antibodies in normal and leukaemic cattle. J. Comp. Path. 86:571-580. 59
Tsukiyama, K., 1985. Control of expression of genome: Effect of plasma blocking factor and platelet derived factor. Jpn. J. Vet. Res. 33:101.
Ungar-Waron, H., Avraham, R., Gluckmann, A. and Trainin, Z., 1978. Reduced immunocompetence of antibodies produced in leukemic cattle. Annales de Recherches Veterinaires 9:815-819.
Van Der Maaten, M.J., Boothe, A.D. and Seger, C.L, 1972. Isolation of a virus from cattle with persistent lymphcytosis. J. Natl. Cancer. Inst. 49:1649-1657.
Van Der Maaten, M.J. and Miller, J.M., 1990. Bovine leukosis virus. In: Virus Infections of Vertebrates, Vol 3: Virus infections of ruminants. Elsevier Science Publishers B.V., Amsterdam, p. 419-429.
Van de Velde, H., von Hoegen, I., Kuo, W,, Pames, J.R. and Thielemans, K., 1991. The B-cell surface protein CD72 is the ligand for CDS. Nature 351:662-665.
Villouta, G., Botto, G. and Rudolph, W., 1988. Spontaneous and mitogen-induced RNA synthesis by blood lymphocytes from bovine leukemia virus-infected and nomnal cows. Vet. Immunol. Immunopathol. 18:287-291.
Weiland, F. and Ueberschar, S., 1976. Ultrastructural comparison of bovine leukemia virus (BLV) with C-type particles of other species. Arch. Virol. 52:187-190.
Weiss, R.A., 1993. How does HIV cause AIDS? Science 260:1273-1278.
Wemer-Fabre, C., Visher, T.L., Wohlwend, D. and Zubler, R., 1989. Cell surface antigen CD5 is a marker for activated human B cells. Eur. J. Immunol. 19:1209-13.
Werling, D., Sileghem, M., Lutz, H. and Langhans, W., 1995. Effect of bovine leukemia virus infection on bovine peripheral blood monocyte responsiveness to lipopolysaccharide stimulation in vivo. Vet. Immunol. Immunopathol. 48:77-88.
Whetstone, C.A., Van Der Maaten, M.J. and Black, J.W., 1990. Humoral Immune response to the bovine immunodeficiency-like virus in experimentally and naturally infected cattle. J. ViroL 64:3557-3561.
Whetstone, C.A., Van Der Maaten, M.J., and Miller, J.M., 1991. A western blot assay for the detection of antibodies to bovine immunodeficiency-like virus in experimentally inoculated cattle, sheep, and goats. Arch. Virol. 116:119-131.
Willems, L, Kettmann, R., Chen, G., Portetelle, D., Bumy, A. and Derse, D., 1992. A cyclic AMP-responsive DNA-binding protein (CREB2) is a cellular transactivator of the bovine leukemia virus long terminal repeat. J. Virol. 66:766-772. 60
Willems, L., Kerkhofs, P., Bumy, A., Mammerickx, M. and Kettmann, R., 1995. Lack of LTR and ENV genetic variation during bovine leukemia virus-induced leukemogenesis. Virology 206:769-772.
Williams, D.L, Amborski, G.F. and Davis, W.C., 1988a. Enumeration of T and B lymphocytes in bovine leukemia virus-infected cattle, using monclonal antibodies. Am J. Vet. Res. 7:1098-1103.
Williams, D.L, Barta, O. and Amborski, G.F., 1988b. Molecular studies of T lymphocytes from cattle infected with bovine leukemia virus. Vet Immunol. Immunopathol. 19:307-323.
Wu, M.C„ Shanks, R.D. and Lewin, H.A., 1989. Milk and fat production in dairy cattle influenced by advanced subclinical bovine leukemia virus infection. Proc. Natl. Acad. Sci. 86:993-996.
Yamamoto, S., Onuma, M., Kodama, H., Mikami, T. and Izawa, H., 1984. Suppression of natural cytotoxic activity of lymphocytes from cattle and sheep during the progress of bovine leukosis. Vet. Micro. 9:105-111.
Zandomeni, R.O., M. Carrera-Zandomeni, M., E. Esteban, E. and J. F. Fen-er, J.F., 1992. Induction and Inhibition of bovine leukemia virus expression in naturally infected cells. J. Gen. Virol, 73:1915-1924. 61
CHAPTER 3. LOSS OF GAG-SPECIFIC ANTIBODY REACTIVITY IN CATTLE EXPERIMENTALLY INFECTED WITH BOVINE IMMUNODEFICIENCY-LIKE VIRUS
A paper published in Viral Immunology, Volume 8, p. 27-36, 1995
Jeffrey A. Isaacson, James A. Roth, Charles Wood, and Susan Carpenter
Abstract
The development and persistence of virus-specific antibodies were investigated In eight cattle experimentally Infected with the R29 isolate of bovine immunodeflcienoy-like vims (BIV). By four weeks post-inoculation (p.i.), antibodies reactive to BIV gag- and env-encoded recombinant fusion proteins were detectable by immunoblotting in all animals. By 40 weeks p.i., seven of eight cattle had dramatically decreased Gag-specific antibodies, and anti-Gag reactivity remained very low or undetectable through 190 weeks p.i. Immunoprecipitation experiments revealed a similar loss of reactivity to non-denatured BIV Gag in these animals. In contrast, antibodies to a recombinant BIV Env protein were readily detectable throughout the study in all eight cattle. During the period of declining Gag antibody, infectious virus was recoverable from peripheral blood mononuclear cells of each animal. However, there was no evidence for sufficient amounts of BIV p26- containing Immune complexes to explain the loss of anti-Gag reactivity.
Interestingly, the single animal which maintained detectable anti-Gag reactivity throughout the study was repeatedly negative for virus recovery beyond 17 weeks 62 p.i. All animals have remained clinically normal for over four years p.i., with no evidence of consistent changes in mononuclear cell subsets. These findings provide evidence that in BIV infection an early decline in Gag-specific antibody reactivity can occur without evidence of increasing viral replication or progression to overt clinical disease.
Introduction
Lentiviral infections are typically characterized by a long incubation period followed by a slow, progressive disease course in the naturally infected host.
However, there are wide variations in the clinical manifestations induced by lentiviruses of different species. Bovine immunodeficiency-like virus (BIV), a lentivirus of cattle, was originally isolated from a dairy cow with persistent lymphocytosis, lymphoid hyperplasia, and perivascular cuffing in the brain (34).
Experimental infection of calves with BIV is associated with a transient increase in lymphocytes (9,32,34) and a lymphoid hyperplasia similar to that found early after infection with immunosuppressive lentiviruses (9,34). Alterations in bovine lymphocyte, monocyte, and neutrophil functions have been reported (14,22,25); however, there has been no indication of BlV-induced immmunosuppressive disease in cattle experimentally infected with either the original R29 isolate of BIV
(9,34) or with recently described field isolates (32). Characterization of virus and host factors which contribute to this clinical quiescence in BlV-infected cattle may further our understanding of lentiviral pathogenesis in other species. 63
Considerable attention has been given to the association between the humoral response to specific human immunodeficiency virus (HIV-1) proteins and development of the acquired immune deficiency syndrome (AIDS). Early after HIV infection, predominant antibody reactivity is directed against the gag gene- encoded p24 capsid protein (16). Overtime, seroreactivity to core proteins often declines in association with disease progression, while anti-envelope antibodies usually persist throughout the course of Infection (1,3,5,7,10,17,18,21,27,29).
Although pattems of antibody reactivity to specific viral antigens are not as well characterized in lentiviral infections of domestic animals, there is some evidence for a similar link between a loss of antibodies to viral core antigens and disease development in feline immunodeficiency virus (FIV) (12,15) and visna virus (19) infections.
Previous studies reported that antibodies to p26, the major BIV capsid protein (4), persist for more than two years p.i. in experimentally infected cattle
(36,37). However, data indicated that two out of the four animals tested at 19-38 months p.i. had lost detectable antibody, as determined by Westem blotting and
IFA (36). The recent development of BIV recombinant proteins (2,11) has provided an abundant source of antigens, which now allows a more extensive detennination of the virus-specific antibody response over the course of BIV infection. These reagents are used in the present study to examine pattems of BIV Gag and Env- specific antibody reactivity in cattle that have been infected with BIV for over four years. This information was compared with virus recovery data and clinical findings in order to better define the role of virus-specific antibodies in controlling virus replication and potential disease development. Though initially strong. Gag- 64 specific antibody reactivity declined dranfiaticaliy in most cattle in the first year following BIV inoculation. Env-specific antibodies, however, were easily detectable in all animals throughout the study. In contrast to other lentiviral infections, the decline in reactivity to BIV Gag proteins was not associated with evidence of increasing viral replication, nor with any overt disease. This suggests that an increase in viral activity is an unlikely explanation for the loss of Gag-specific antibodies in BIV infection.
Materials and Methods
Animal inoculation, serum collection, and clinical monitoring
Colostrum-deprived male Holstein calves, two to four months of age, were purchased from a local dairy and were serologically determined to be free of antibody to BIV, bovine leukemia virus (BLV), and bovine diarrhea virus (BVDV).
Calves were vaccinated with killed BVDV vaccine (EXCEL 4; Bio-Ceutic, St.
Joseph, MO) as previously described (9) and eight calves were inoculated intravenously with 1.8 X 10^^ syncytia fomning units of BIV R29 propagated in fetal bovine lung cells (FBL) (9). Five of the eight calves received 1 x 10^ peripheral blood mononuclear cells (PBMC) from a lymphocytotic cow persistently infected with BLV as part of a separate study on the interactive effects of the two viruses.
The other three calves also received 10^ PBMC from a BLV-free calf. Three control calves were inoculated with BlV-free FBL and normal PBMC. Groups were housed in separate pens, and monitored daily for external signs of clinical disease. Blood 65 was collected by jugular venipuncture on a weekly basis for the first 11 weeks post-
Inoculation (p.i.). then every four to twelve weeks thereafter through 190 weeks p.i.
Total leukocytes were counted and mononuclear cell subset percentages were determined by flow cytometry at similar internals as part of another study (K.P.
Flaming, manuscript in preparation).
Virus isolation
PBMC were isolated from whole blood by centrifugation, collection of the buffy coat, and hypotonic lysis of erythrocytes (28). Approximately 10^ PBMC were co-cultivated with primary FBL (9) in the presence of eight ^ig/ml polybrene.
Cultures were passaged twice weekly and visually monitored for syncytium formation. At each passage, duplicate cultures were assayed for the presence of
BIV, BLV, and bovine syncytium-forming virus using immunofluorescence or immunoperoxidase assays (9,35). Serologic reagents used for detection of BIV included convalescent serum from a calf experimentally infected with BIV (9), and
BIV p26-specific monoclonal antibodies (35). For detection of BLV, high-titered antisera was collected from a sheep experimentally inoculated with PBMC from a
BLV-infected cow with persistent lymphocytosis. In most cases, cells were sub- cultured at least three times before they were considered virus negative. 66
BIV antigens and immunologic reagents
The cloning and expression of BIV recombinant TrpE fusion proteins has
been reported (2,11,35,38). The clone referred to here as BIV gag-3 contains the
entire p26 coding region of BIV and small flanking regions on the 3' and 5' ends,
ligated Into a pATH expression vector. The env-8 clone similarly expresses a
fusion protein containing 134 amino acids from the amino terminal portion of the
BIV transmembrane envelope protein gp42, which appears to contain a major
immunoreactlve epitope of BIV (11). Following induction of protein expression, the bacteria were lysed, and recombinant proteins (Gag3 or Env8) were precipitated,
washed, dissolved in 0.5% SDS, and stored at -70°C.
To obtain hyperimmune antisera to BIV p26, rabbits were immunized intramusculariy with 100 ^.g Gag3 (1 mg/ml in 0.5% SDS) every two to three weeks for a total of five doses. The first dose was administered in an equal volume of a microparticulate stabilized water-in-oil squalene emulsion (Titer Max; Vaxcel, Inc.,
Norcross, GA); remaining doses consisted of Gag3 dissolved in 0.5% SDS.
Beginning after the second injection, blood samples were screened for reactivity to
Gag3 and BIV p26 by Western immunoblotting. BIV p26-specific monoclonal
antibodies were generated against Gag3 as reported (35). Hybridoma
supematants were screened on the basis of reactivity to p26 from purified BIV.
MAb #104 recognized both the 65 kDa Gag3 protein and a smaller band of about
54 kDa, which was not recognized by some other p26-specific MAbs (35). 67
Western immunobiotting
Aliquots containing 30 jig of Gag3 only or 30 ng each of Gag3 and Env8 were resolved by SDS-PAGE on 10% preparative gels at 200V for 35 minutes in a
Bio-Rad (Richmond, CA) mini-Protean system. Proteins were transfen-ed to nitrocellulose membranes at 100V for 60 minutes using a Bio-Rad transblot system.
Membranes were blocked ovemight at 4°C in 1% gelatin in TST (lOmM Tris, 150 mM NaCI, 0.1% Tween 20). Washed membranes were placed in a Miniblotter 16 or 28 apparatus (Immunetics, Cambridge, MA), incubated with 1:20 test sera at room temperature for 2 hours, washed again, and incubated for 1 hour with 0.06 jiCi/ml of I^Sf-iabeled protein G (Amersham, Arilngton Heights, IL). Membranes were air-dried and placed on radiographic film for 12-48 hours at -70°C.
Immunoprecipitation of GagS
After patterns of antibody reactivity to BIV antigens were determined by immunobiotting, immunopreciprtations were performed to detect antibodies specific for conformational epitopes of non-reduced Gag3. To minimize non-specific binding, the Gag3 preparation was pre-cleared by ovemight incubation with protein
G-agarose beads (Boehringer Mannheim, Indianapolis, IN). Twenty ^il test serum, alone or mixed with 0.5 ng pre-cleared GagS, was brought to 200 nl volume In lysis buffer (8) containing 1% Nonidet P-40, 0.5% deoxycholate, and 0.1% SDS, and
Incubated 2 hours with rotation at room temperature. Ten fig protein G-agarose beads was then added to precipitate IgG Immune complexes. After 2 hours, beads 68 were pelleted by brief centrifugation, washed three times in 1 ml lysis buffer, and boiled to separate bound antigen from the complexes. Immunoprecipitated antigens were detected by immunoblotting using polyclonal rabbit anti-Gag3 antibody diluted 1:500, followed by incubation with protein G as stated above.
Results
Virus recovery and clinical observations
BIV was readily recovered from PBMC of all but one animal on four separate occasions during the first 24 weeks p.i. (Table 1). Only two cattle, #342 and 345, were BlV-positive at every isolation attempt during the period of study. Animal
#348 was BlV-negative on only one occasion, and #341 and 347 tested negative twice. Animal #344, however, was unique in remaining consistently BlV-negative after week 17 of the study. There was no apparent difference in pattems of BIV recovery from PBMC of animals inoculated with BIV alone (#340, 341, 342) and those co-infected with BLV (#344, 345, 347, 348, 349).
Previous studies have indicated that during the first six weeks p.i., mean lymphocyte counts were significantly higher in the BlV-inoculated animals compared to uninfected control calves (9). However, from ten weeks through four years p.i., there were no consistent BlV-induced changes in numbers of total leukocytes, GD4+ or CD8+ T cells, B cells, or monocytes (K. P. Flaming, manuscript in preparation). In addition, no lymphadenopathy or other clinical abnormalities 69 attributable to BIV were noted in any of these cattle through four years p.i. One BIV-
Infected animal developed fatal lymphosarcoma at six months p.i. and was not included in the present study.infected animal developed fatal lymphosarcoma at six months p.i. and was not included in the present study.
Table 1. Recovery of virus from PBMC of cattle experimentally infected with BIV or with BIV and BLV
Virus Recoveryc-weeks post-inoculation
Animal Inoculumb 2 6 16 27 52 75 95 112 143 166 190
340a BIV + + + + - - + + - - -
341 BIV + + + + - - + + + + +
342 BIV + + + + + + + + + + +
344 BIV + + BLV + + + + + + + + + + +
345 BIV + + + + + + + + + + BLV + + + + + + + + + + +
347 BIV + + + + + + + + + BLV + + + + - + - - + + -
348 BIV + + + + + + + + + + BLV + + + + + + + + + + +
349 BIV + + + + _ + + + BLV + + + + + — — + + +
®This animal is persistenth/ infected with bovine syncyta virus ''Bght calves were inoculated intravenously with 1.8 X10^ syncytia fomiing units of BIV R29 in fetal bovine lung cells; five also received 10^ PBMC from a lymphocytotic cow persistently infected with BLV. ^irus was isolated by coculturing 10^ PBMC with FBL cells and monitoring for syncytia formation; viruses were differentiated by immunocytochemistry. Cultures were passed at least three times before they were considered negative. 70
Patterns of antibody reactivity to BIV Gag3 recombinant protein by immunoblotting
Sequential serum samples from cattle experimentally infected with BIV R29 exhibited detectable IgG reactivity to recombinant GagS by four weeks p.i., as exemplified by animal #342 (Fig. 1), In seven of eight BlV-infected cattle, antibody reactivity to Gag3 began to decrease between six and ten weeks after infection, and diminished to low or undetectable levels by week 40. This loss of anti-Gag reactivity was not associated with any predictable changes in viral activity. For example, although Gag-specific antibodies were lost over a similar period of time in all seven cattle, virus was recoverable from the PBMC of animals #342 and #345 throughout the study, while recovery from animals #340, #341, and #349 was less consistent (Table 1). In most instances, two to three blind passages were required to detect BIV replication by immunocytochemistry. Additional experiments demonstrated that there was not a linear relationship between the number of virus- infected cells and the number of passages required for detection of virus replication
(not shown). Therefore, it was not possible to quantitatively compare BlV-infected
PBMC with the antibody titer to Gag and/or Env. Similarly, the antibody reactivity detectable by Westem blotting is not strictly quantitiative, but rather represents an overall decline in Gag-specific antibodies.
A different pattern of antibody reactivity was seen with sera from calf #344.
Anti-Gag reactivity in this animal remained easily detectable through at least 166 weeks p.i. (Fig. 1 & 2). In addition, reactivity to the BlV-specific 54-kDa band 71 appeared to be more pronounced in this animal compared to others, indicating that different or additional p26-specific epitopes may have been recognized.
Interestingly, this prolonged anti-Gag reactivity was observed only in the single animal which remained consistently negative for virus recovery beyond week 27.
Weeks post-BIV inoculation
0 2 4 6 10 16 27 40 60 75 95 + Mab 0 2 4 6 10 16 27 40 60 75 95 + Mab 200- 97.4-
Calf#342 Calf #344
Figure 1. Seroreactivity to the 65 kDa Gag3 recombinant fusion protein in cattle experimentally infected with BIV. All serum samples were diluted 1 ;20 in TST buffer for testing. Positive controls consisted of a 1:20 dilution of convalescent bovine serum collected 40 days post-BIV inoculation (+) (10), along with undiluted BIV p26-specific monoclonal antibody #104 (MAb) (35). Seven of the eight BIV- infected animals exhibited reactivity patterns very similar to animal #342, but only animal #344 had persistent reactivity to GagS throughout the study. There was no comparable reactivity to lysates from bacteria transfomied with pATH alone (data not shown): thus, reactivity was specific for the BIV gag-encoded portion of the fusion protein. Bands of approximately 40 and 46 kDa presumably represent varying reactivity to non-BlV E. coli proteins that remain in the antigen preparations. 72
Antibody reactivity to Env8
A BIV env-encoded recombinant fusion protein was used to further characterize the specificity of the antibody loss. Seroreactivity to Env8 was readily detectable throughout the study in all eight cattle, as shown for animals #341,
#342, #344, and #349 (Fig. 2). Sera from all cattle continued to react to the Env8 protein for over four years p.i. (data not shown). The persistence of reactivity to
Env8 in Gag-seronegative cattle provides evidence for the Gag-specific nature of the early antibody loss. It should also be noted that the presence of BLV co- infection appeared to have no consistent effect on the humoral response to BIV gag or env^-encoded proteins.
Immunoprecipitation of Gag3
To mle out the possibility that antibody reactivity shifted from linear to confomnational Gag epitopes over time, immunoprecipitation experiments were conducted using non-reduced Gag3. As shown for a representative animal in Fig.
3 ("+" lanes), antibodies from all BlV-infected cattle precipitated non-denatured
Gag3 by ten weeks p.i., with reactivity becoming greatly diminished by 60 weeks p.i. in seven out of eight animals. Moreover, no p26 bands were detected in the lanes (where no Gag3 was added), even in serum samples which had lost anti-
Gag reactivity. If p26 antigenemia was present in sufficient amounts to block Gag3 antibodies, a p26 band of similar density to the precipitated Gag3 would have been expected to appear as reactivity to Gag3 diminished. This provides evidence that Figure. 2. Persistence of antibody reactivity to a BiV env-encoded protein. Equivalent amounts of Gag3 and Env8 recombinant antigens were pre-mixed and resolved by preparative SDS-PAGE, followed by Western immunoblotting. Serum dilutions and positive controls are the same as described for Figure 1. Shown are representative examples of animals infected with BIV only (#341 and #342) and with BIV and BLV (#344 and #349). Weeks post-BIV inoculation
^. SQ.CS|<0^
133 —
71 — f Qag3 t • -4 Erw8 41.8—! 30.6— #341 #344
133 — I, f ^ Qafl3 71 — m< EnvS % 41.8—^
30.6— #342 #349 75 the formation of p26-containing circulating immune complexes (CIC) does not account for the loss of anti-Gag reactivity. However, if CIC were rapidly cleared from the circulation, the assay used here might not have been sensitive enough to detect low levels remaining in the serum.
Weeks post-BiV inoculation
CO 0 4 10 60 95 112 > ^ SO- + - + - +- + -+- +
97.4— A ^ •-Gag3 46- — *
30—! m ^p26
Figure. 3. Immunoprecipitation of IgG immune complexes from BlV-infected cattle. Test sera were diluted 1:10 in the presence ("+" lanes) or absence lanes) of Gag3, then incubated with protein G-agarose beads and assayed for complexed GagS or BIV p26 by Western immunoblotting. The"+" lanes, therefore, would show the presence of antibodies capable of immunoprecipitating the non-reduced fomn of GagS. Seven of the BlV-infected cattle had patterns of reactivity similar to that shown for animal #342; while only animal #344 had prolonged reactivity. No p26 was detected In any of the lanes, including those samples that had lost reactivity to Gag3. Positive control lanes (left) contain^ 10 micrograms of pelleted BIV or five micrograms of Gag3. •'25|-iabeled Protein G did not react with heavy chains of denatured bovine IgG transferred directly onto nitrocellulose membranes (data not shown). 76
Discussion
In this study, longitudinal patterns of antibody reactivity to recombinant BIV proteins were determined in eight experimentally infected cattle, all of which have remained clinically healthy for over four years p.i. By four weeks p.i., strong antibody reactivity to BIV gag- and env-encoded recombinant fusion proteins was detectable by immunoblotting in sera from all animals. However, reactivity to the
BIV Gag3 recombinant protein became greatly decreased or undetectable by 40 weeks p.i. in seven cattle. Immunoprecipitation experiments revealed a similariy diminished reactivity to Gag3 confomnational epitopes by 60 weeks p.i. In contrast, reactivity to the Env8 recombinant protein was readily detectable in all BlV-infected cattle throughout the four-year study period, suggesting that the observed phenomenon specifically affected Gag-specific antibodies. Virus was recovered from the PBMC of each animal several times before and after the loss of Gag- specific antibodies, and there was no obvious temporal association between the period of antibody loss and changes in virus recovery. Previous studies of serological reactivity in infections with other Antiviruses such as HIV, FIV, and visna virus have found a relationship between the loss of antibodies to viral core proteins and the development of clinical disease (1,5,10,12,15,17,18,19,21,27,29).
In contrast, BIV infection has not been associated with disease, and we observed no signs of deteriorating health or consistent changes in lymphocyte subset counts in any cattle which had lost antibody to BIV core proteins. While it is possible that
BlV-induced disease may yet develop in these animals, it is more likely that the relatively early loss of BIV Gag-specific antibodies occurs independently of 77 progression to overt disease. This observation suggests that alternative mechanisms may be responsible for the loss of Gag-specific antibodies in BiV
Infection in comparison to other lentiviral infections.
In HIV infection, it is postulated that increasing virus replication and p24 antigen production can cause immune complex fonnation with concomitant loss of detectable antibody reactivity to p24 (1,7,17,21). However, the loss of antibodies to
HIV core proteins has also been observed in the absence of detectable antigenemia or clinical symptoms, leading some investigators to hypothesize that other mechanisms may also play a role in this phenomenon (6,13,24,26). Indeed, it has been reported that PBMC from HIV p24 antibody-negative patients synthesize less Gag-specific antibody than p24 seropositive individuals (30,33). In the case of BIV, several lines of evidence argue against an Immune complex- dependent mechanism as sufficient explanation for the loss of Gag-specific antibodies in Infected cattle. First, an immunoprecipltation assay revealed no direct evidence of p26-contalning immune complexes in any of the seven Gag- seronegative cattle. In addition, there was no increase in virus recovery from the
PBMC of these animals over time, as might have been expected if increasing viral replication accompanied the loss of Gag seroreactivity. Finally, no signs of immune complex disease have been obsen/ed in the Gag-seronegative animals. While it is possible that circulating immune complexes could have been removed from the blood by the mononuclear phagocytic system soon after their formation, it is unlikely that these animals would have remained clinically unaffected by prolonged high levels of virus replication and immune complex formation. 78
In view of this evidence, it is likely that immune complex-independent mechanisms are involved in the decline of BIV Gag-specific antibodies. Genetic differences in immune responsiveness to BIV Gag may account for some animal to animal variability, but are unlikely to account for the antibody loss in seven of eight animals. The maintenance of strong anti-Env reactivity in all eight cattle, along with virus recovery data, suggests that the loss of Gag-specific antibodies is not simply attributable to a functional latency of BIV in vivo. In fact, virus was most difficult to recover from the single animal which maintained antibodies to Gag throughout the study. One interpretation of these findings is that the presence of antibodies directed against certain Gag-specific epitopes might exert a protective effect against viral replication. However, the loss of anti-Gag reactivity over time did not lead to an increase in virus recovery in the other seven animals, as might have been expected if the presence of Gag-specific antibodies was protective.
Continuing work should focus on the further elucidation of the mechanism(s) behind the loss of Gag-specific antibodies, which would increase our understanding of host-virus interactions during periods of clinical quiescence in other lentiviral infections.
Lastly, these findings raise concern regarding serodiagnosis of BIV in naturally infected animals. In this study, seven out of eight BlV-infected cattle consistently lost detectable reactivity to either linear or conformational BIV Gag epitopes within 1 year p.i. Using antigens prepared from BlV-infected cell culture,
Suarez et al recently reported a similar decline in anti-p26 antibodies in two cattle infected with either BIV R29-1203 or the recently described field isolate FL491 (31).
Assays that rely on detection of anti-Gag antibody might therefore lead to an 79
underestimation of BIV infection rates, and a misunderstanding of the potential role
of BIV in bovine health problems. The more prolonged seroreactivity to the Env8
recombinant protein suggests that it would be a useful antigen for incorporation into BIV diagnostic assays. Reliable detection of BIV infection might also be achieved by PGR amplification of BlV-specific DNA (20,23,31,32). The use of more sensitive assays for detection of BIV infections will allow better epidemiologic studies to determine whether a loss of Gag-specific antibody reactivity is a common phenomenon in natural BIV infections.
Acknowledgements
We thank Andrea Dom, Jim Fosse, Thomas Skadow, and Yvonne
Wannemuehler for technical assistance, and Willie Parkhurst, Don Robinson, and
Virgil Schultz for animal care. This work was supported by PHS grants R01
CA50159, R01 GA59125, and T32 AI07378 to JAR and CA62810 and AI30356 to
CW.
References
1. Allain, J.P., Y. Laurian, A. Paul, F. Verrouse, M. Leuther, G.,Gazengel, D. Senn, M.J. Larrieu and G. Bossner. 1987. Long-term evaluation of HIV antigen and antibodies to p24 and gp41 in patients with hemophilia: potential clinical importance. N. Engl. J. Med. 317:114-121.
2. Atkinson, B., Z.Q. Liu and G. Wood. 1992. Use of bacterial trpE fusion vectors to express and characterize the bovine immunodeficiency-like virus core protein. J. Virol. Methods 36:35-49. 80
3. Barin, F., M.F. McLane, J.S. Allsin, T.H. Lee, J.E. Groopman and M. Essex. 1985. Virus envelope protein of HTLV-lll represents major target antigen for antibodies in AIDS patients. Science 228:1094-1096.
4. Battles, J.K., M.Y. Hu, L Rasmussen, G.J. Tobin and M.A. Gonda. 1992. Immunological characterization of the gag gene products of bovine immunodeficiency virus. J. Virol. 66:6868-6877.
5. Biggar, R.J., M. Melbye, P. Ebbesen, S. Alexander, J.O. Nielsen, P. Sarin and V. Faber. 1985. Variation in human T lymphotropic virus III (HTLV-lll) antibodies in homosexual men: decline before onset of illness related to acquired immune deficiency syndrome. Br. Med. J. 291:997-998.
6. Burger, H., B. Weiser, W.S. Robinson, J. Lifson, E. Engleman, C. Rouzloux, R., Bmn-Vezinet, F. Barre-Slnoussi, L. Montagnler and J.C. Chermann. 1985. Transient antibody to lymphadenopathy-associated virus/human T lymphotropic virus type III and T-lymphocyte abnormalities in the wife of a man who developed the acquired immunodeficiency syndrome. Ann. Intem. Med. 103:545-547.
7. Cao, Y., F. Valentine, S, Hojvat, J.P. Allain, P. Rubenstein, M. Mirabile, S. Czelusniak, M. Leuther, L. Baker and A.E. Friedman-Kien. 1987. Detection of HIV antigen and specific antibodies to HIV core and envelope proteins In sera of patients with HIV infection. Blood 70:575-578.
8. Carpenter, S., L.H. Evans, M. Sevoian and B. Chesebro. 1987. Roie of the host immune response in selection of equine infectious anemia virus variants. J. Virol. 61:3783-3789.
9. Carpenter, S., L.D. Miller, S. Alexandersen, C.A. Whetstone, M.J. VanDerMaaten, B. Viuff, Y. Wannemuehler, J.M. Miller and J.A. Roth. 1992. Characterization of early pathogenic effects after experimental Infection of calves with bovine immunodeficiency-like virus. J. Virol. 66:1074-1083.
10. Cheingsong-Popov, R., C. Panagiotidi, S. Bowcock, A. Aronstam, J. Wadsworth and J. Weber. 1991. Relation between humoral responses to HIV gag and env proteins at seroconversion and clinical outcome of HIV infection. Br. Med. J. 302:23-26.
11. Chen, P., Z.Q. Liu and C. Wood. 1994. Use of TipE fusion protein to identify antigenic domains within the BIV envelope protein. J. Virol. Methods 47:331- 343.
12. Egberink, H.F., C.E.J.M. Keldermans, M.J.M. Koolen and M.C. Hon/inek. 1992. Humoral immune response to feline immunodeficiency virus in cats with experimentally induced and naturally acquired infections. Am. J. Vet. Res. 53:1133-1138. 81
13. Farzadegan, H., M.A. Polis, S.M. Wolinsky, C.R. Rinaldo.Jr., J.J. Sninsky, S. Kwok, R.L Griffith, R.A. Kaslow, J.P. Phair, B.F. Polk and A.J. Saah. 1988. Loss of human immunodeficiency virus type 1 (HlV-1) antibodies with evidence of viral infection in asymptomatic homosexual men. Ann. Intern. Med. 108:785- 790.
14. Flaming, K., M. VanDerMaaten, C. Whetstone, S. Caipenter, D. Frank and J. Roth. 1993. The effect of bovine immunodeficiency-like virus infection on immune function In experimentally infected cattle. Vet. Immunol. Immunopathol. 36:91-105.
15. Furuya, T., A. Hasegawa, T. Miyazawa, K. Miki and T. Mikami. 1992. Detection of anti-gag antibodies of feline immunodeficiency virus in cat sera by enzyme linked immunosorbent assay. Arch. Virol. 124:355-361.
16. Gaines, H., M. von Sydow, A. SonneriDorg, J. Albert, J. Czajkowski, P.O. Pehrson, F. Chiodi, L. Moberg, E.M. Fenyo, 8. Asjo and M. Forsgren. 1987. Antibody response in primary human immunodeficiency virus infection. Lancet ii:1249-1253.
17. Goudsmit, J.J., M.A. Lange, D.A. Paul and G.J. Dawson. 1987. Antigenemia and antibody titers to core and envelope antigens in AIDS, AIDS-related complex, and subclinical human immunodeficiency virus infection. J. Infect. Dis. 155:558-560.
18. Groopman, J.E., P.I. Hartzband, L. Shulman, S.Z. Salahuddin, M.G. Samgadharan, M.F. McLane, M. Essex and R. Gallo. 1985. Antibody seronegative human T-lymphotropic virus type III (HTLV-lll)-infected patients with acquired immunodeficiency syndrome or related disorders. Blood 66(3):742-744.
19. Houwers, D.J. and I.M. Nauta. 1989. Immunoblot analysis of the antibody response to ovine lentivirus infections. Vet. Microbiol. 19:127-139.
20. Kashanchi, F., Z.Q. Liu, and C. Wood. 1991. Comparative evaluation of bovine immunodeficiency-like virus infection by reverse transcriptase and polymerase chain reaction. J. Virol. Methods 31:197-210.
21. Lange, J.M.A., D.A. Paul, H.G. Huisman, F. de Wolf, H. van den Berg, R.A. Coutinho, S.A. Danner, J. van der Noorda and J. Goudsmit. 1986. Persistent HIV antigenaemia and decline of HIV core antibodies associated with transition to AIDS. Br. Med. J. 293:1459-1462.
22. Martin, S.J., TP. O'Neill, J.A. Bilello and J.L. Eiseman. 1991. Lymphocyte transformation abnormalities in bovine immunodeficiency-like virus infected calves. Immunol. Lett. 27:81-84. 82
23. Nadin-Davis, S., S.C. Chang, H. Smith and R.M. Jacobs. 1993. Detection of bovine immunodeficiency-lllce virus by the polymerase chain reaction. J. Virol. Methods 42:323-336.
24. Nishanian, P., K.R. Huskins, S. Stehn, R. Detels and J.L Fahey. 1990. A simple method for improved assay demonstrates that HIV p24 antigen is present as immune complexes in most sera from HIV-infected individuals. J. Infect. Dis. 162:21-28.
25. Onuma, M., E. Koomoto, H. Furuyama, Y. Yasutomi, H. Taniyama, H. Iwai and Y. Kawakami. 1992. Infection and dysfunction of monocytes induced by experimental inoculation of calves with bovine immunodeficiency-like virus. JAIDS 5:1009-1015.
26. Portera, M., F. Vitale, R. La Licata, D.R. Alesi, G. Lupo, F. Bonura, N. Romano and G. Di Cuonzo. 1990. Free and antibody-complexed antigen and antibody profile in apparently healthy HIV seropositive individuals and in AIDS patients. J. Med. Virol. 30:30-35.
27. Ragni, M.V., T. O'Brien, D. Reed, J.A. Spero and J.H. Lewis. 1988. Prognostic importance of antibodies to human immunodeficiency virus by recombinant assay and Western blot techniques in HIV antibody-positive hemophiliacs. AIDS Res.Hum.Retroviruses 4:223-231.
28. Roth, J.A., M.L. Kaeberie and R.W. Griffith. 1981. Effects of bovine viral diarrhea virus infection on bovine polymorphonuclear leukocyte function. Am. J. Vet. Res. 41:244-250.
29. Schupbach, J., O. Haller, M. Vogt, R. Luthjy, H. Joller, O. Oelz, M. Popovic, M.G. Samgadharan and R.C. Gallo. 1985. Antibodies to HTLV-III in Swiss patients with AIDS and pre-AIDS and in groups at risk for AIDS. N. Engl. J. Med. 312:265-270.
30. Stickler, M.C., S. Shaipe, and S. Zolla-Pazner. 1992. p24 antibody production in p24 antibody-negative HIV-infected subjects. Viral Immunology 5:123-32.
31. Suarez, D.L., M.J. VanDerMaaten and C.A. Whetstone. 1995. Improved early and long-temn detection of bovine lentivirus by a nested polymerase chain reaction test in experimentally infected calves. Am. J. Vet. Res. 56:579-86.
32. Suarez, D.L., M.J. VanDerMaaten, C. Wood and C.A. Whetstone. 1993. Isolation and characterization of new wild-type isolates of bovine lentivirus. J. Virol. 67:5051-5055.
33. Teeuwsen, V.J.P., J.M.A. Lange, R. Keet, J.K.M.E. Schattenkert<, C. Debouck, R. van den Akker, J. Goudsmit, and A.D.M.E. Osterhaus. 1991. Low numbers of functionally active B lymphocytes in the peripheral blood of HIV-1 - 83
seropositive individuals with low p24-specific semm antobody titers. AIDS 5:971-979.
34. VanDerMaaten, M.J., A.D. Boothe and C.L Seger, 1972. Isolation of a virus from cattle with persistent lymphcytosis. J. Natl. Cancer. Inst. 49:1649-1657.
35. Wannemuehler, Y., J. Isaacson, M. Wannemuehler, C. Wood, J.A. Roth and S. Carpenter. 1993. In vitro detection of bovine immunodeficiency-like virus using monoclonal antibodies generated to a recombinant gag fusion protein. J. Virol. Methods 44:117-128.
36. Whetstone, C.A., M.J. VanDerMaaten and J.W. Black. 1990. Humoral immune response to the bovine immunodeficiency-like virus in experimentally and naturally infected cattle. J. Virol. 64:3557-3561.
37. Whetstone, C.A., M.J. VanDerMaaten and J.M. Miller. 1991. A westem blot assay for the detection of antibodies to bovine immunodeficiency-like virus in experimentally Inoculated cattle, sheep, and goats. Arch. Virol. 116:119-131.
38. Windheuser, M.G. and C. Wood. 1988. Characterization of immunoreactive epitopes of the HIV-1 p41 envelope protein using fusion proteins synthesized in Escherichia coll. Gene 64:107-119, 84
CHAPTER 4. EFFECTS OF LONG-TERM INFECTION WITH BOVINE IMMUNODEFICIENCY VIRUS AND/OR BOVINE LEUKEMIA VIRUS ON HUMORAL AND LYMPHOCYTE PROLIFERATIVE RESPONSES IN CATTLE
A paper to be submitted to Veterinary Immunology and Immunopathology
Jeffrey A. Isaacson, Kevan P. Flaming, and James A. Roth
Abstract
Immune responses were examined in cattle experimentally infected with bovine Immunodeficiency virus (BIV) and/or bovine leukemia virus (BLV) for over three years. Lymphocyte proliferative responses to Con A or allogeneic MHO were determined by ^H-thymidine incorporation assays, and antigen-specific humoral and lymphocyte proliferative responses were measured following vaccination with tetanus toxoid (IT) and bovine herpes vims-1 (BHV-1). Lymphocytes from BIV- infected cattle had significantly (p < 0.05) reduced proliferative responses to Con A, but responses to allo-MHC and TT did not differ from those of uninfected controls.
BIV infection also had little effect on TT-specific antibody responses in vivo. In contrast, BLV-infected cattle had significantly increased secondary antibody responses to vaccination with TT, as well as enhancement of antibody responses to BHV-1, Numbers of circulating mononuclear cells were also higher in BLV- infected cattle, which was attributable to Increases In both T and B cell numbers.
Unstimulated lymphocytes from BLV-infected cattle had significantly increased spontaneous uptake of ^H-thymidine in vitro. When differences in counts per 85 minute were analyzed, lymphocytes from BLV-infected cattle had slightly increased proliferative responses to Con A, but no consistent alterations in responsiveness to allo-MHC, TT, or BHV-1. The increased humoral responses to non-BLV antigens in
BLV-infected cattle, without a concomitant increase in antigen-specific lymphocyte proliferation In vitro, suggests that BLV infection may cause a non-specific B cell activation.
Abbreviations
BHV-1, bovine herpes virus-1; BIV, bovine immunodeficiency virus; BLV, bovine leukemia virus; BSV, bovine syncytium-fomriing virus; Con A, concanavalin
A; CPM, counts per minute; DCPM, difference in CPM; ELISA, enzyme-linked immunosorbent assay; HIV, human immunodeficiency virus; HTLV, human T cell leukemia virus; MAb, monoclonal antibody; MHC, major histocompatibility complex;
PBMC, peripheral blood mononuclear cells; PBS, phosphate-buffered saline; p.i., post-inoculation, PL, persistent lymphocytosis; SI, stimulation index; SLP, spontaneous lymphocyte proliferation; TT, tetanus toxoid; WBC, white blood cells.
Introduction
Cattle in the United States are susceptible to persistent infection by at least three distinct retroviruses; bovine syncytial virus (BSV), bovine immunodeficiency virus (BIV), and bovine leukemia virus (BLV). Like the spumavinjses of other species, BSV has not been associated with any bovine health problems (Jacobs at 86 al., 1995). Although BIV is genetically and structurally similar to immunosuppressive lentiviruses such as the human immunodeficiency virus (HIV), there has been no unequivocal evidence of BlV-induced disease in cattle experimentally Infected with either the original R29 isolate of BIV (Van der Maaten et al., 1972, Carpenter et al., 1992) or with recently described field isolates (Suarez et al., 1993). However, BIV infection has recently been associated with the development of an atypical T cell lymphosarcoma in an experimentally Infected calf
(Rovid et al., 1996). BLV is an oncogenic retrovirus that can cause lymphoid tumors and persistent lymphocytosis in its host, although most infected cattle remain clinically and hematologically nomnal (Bumy et al., 1988). Therefore, these three retroviruses appear to cause asymptomatic infections in the majority of infected cattle. The asymptomatic stages of lentiviral and oncoviral infections in other species are often accompanied by altered immune functions. This suggests that BIV and/or BLV might similariy affect the immune responses of cattle in the absence of overt disease.
Previous investigations have suggested that bovine immune functions may be compromised in the early stages of BIV infection. BlV-inoculated calves have been reported to have decreased monocyte functions and delayed humoral responses to protein antigens (Onuma et al., 1992), as well as decreased lymphocyte proliferative responses to mitogens (Martin et al., 1991) In the first year post-inoculation with BIV. However, these conclusions were based on a single sampling date, a single BlV-infected calf, and comparison to a single control animal, respectively, thus their significance is unclear. Studies of larger groups of
BlV-infected cattle and controls over longer periods of time have found evidence of 87
depressed neutrophil functions (Flaming et al., 1993, Flaming et al., In press), but normal monocyte functions (Rovid et al., 1995) and relatively nonnal lymphocyte proliferative responses to mitogens (Flaming et al., in press) through four years p.i.
Considering the ongoing concern about BIV as a factor in bovine health problems
(Brownlie et al., 1994), as well as the potential importance of BIV as a model of HIV infection in humans (Gonda et al., 1990), further investigation of the effects of BIV infection on acquired immune responses in cattle is warranted.
HIV provides a well-studied example of a lentiviral infection with differential effects on the host over time. Much evidence suggests that abnomialities in both T and B cell functions develop long before the loss of CD4+ T cells that coincides with the development of clinical symptoms. Lymphocytes isolated from the peripheral blood of most asymptomatic, HIV-infected individuals have been shown to lose responsiveness to various stimuli as the infection progresses: first to recall antigens, then to allogeneic MHC, and finally to mitogens (Shearer et al., 1986).
Some clinically nonnal, HIV-infected humans also have decreased antibody responses following vaccination with tetanus toxoid or other antigens (Pinching,
1991). SImilariy, feline immunodeficiency virus-infected, asymptomatic cats have been shown to have decreased proliferation of naive T cells to new antigens
(Bishop et al., 1992) as well as to recall antigens (Torten et al., 1991) prior to any change in responses to mitogens or decline in CD4/CD8 ratios. If experimental BIV infection does adversely affect immune functions in outwardly healthy cattle, a reasonable hypothesis is that such manifestations might closely resemble those seen in the asymtomatic stages associated with other lentiviral infections. 88
Although considerably more research effort has been devoted to BLV than to
BIV, the mechanism by which BLV can cause persistent lymphocytosis and/or lymphoid tumors in infected cattle remains unclear. BLV is closely related to the human T cell leukemia viruses, HTLV-I and -II (Sagata et al., 1995). There is evidence that humans with combined HIV and HTLV-II infection can have exacerbated immunosuppressive disease (Beiike et al., 1994). Therefore, it was of interest to detemnine whether co-infection with BIV and BLV would alter the effects of either virus on the bovine immune system. The present study was designed to investigate the effects of long-term infection with these viruses on humoral and cell- mediated immune responses in cattle.
Materials and Methods
Animals and virus inoculations
Colostrum-deprived newbom male Holstein calves were purchased from a local dairy and determined to be free of antibodies to BIV, BLV, and bovine viral diarrhea virus (BVDV). Since the BIV R29 isolate was known to be Infected with non-cytopathic BVDV, all calves were vaccinated with killed BVDV vaccine (EXCEL
4; Bio-Ceutic, St. Joseph, MO) as previously described (Carpenter et al., 1992). At four to six months of age, animals were randomly divided into four groups: four calves were inoculated intravenously with 1.8 X 10^ syncytia fomning units of BIV
R29 propagated in fetal bovine lung cells (FBL); five calves received 1 x 10^ peripheral blood mononuclear cells (PBMC) from a lymphocytotic cow persistently infected with BLV, but negative for BIV and BSV; and five calves received both BIV and BLV. The BlV-only calves also received 1 x 10^ PBMC from a BLV-free calf, and the BLV-only calves also received BlV-free FBL. Three control calves were sham-inoculated with normal PBMC and BlV-free FBL. A noncytopathic isolate of
BVDV was included in the FBL inocula to control for the effects of BVDV in the BIV inocula (Roth et al., 1981). Groups were housed in separate pens and monitored daily for extemal signs of clinical disease.
Peripheral blood cell counts
Total and differential white blood cell (WBC) counts were monitored in BLV- infected and uninfected cattle during the study period. Total WBC counts were determined on whole blood using an electronic particle counter, and differential counts were performed on Wright-stained whole blood smears. For analysis of phenotypic markers in freshly Isolated mononuclear cell subsets, PBMC were isolated as previously described (Flaming et al, 1993). Briefly, venous blood was collected into EDTA tubes, centrifuged, and buffy coats were diluted 1:1 in PBS and layered over Histopaque 1077 (Sigma). Following centrifugation at 400 x g for 30 minutes, mononuclear cell layers were removed, washed, counted, and resuspended in PBS++ (PBS with 0.5% BSA and 0.1% sodium azide to stabalize membrane proteins) for monoclonal antibody (MAb) staining. Cells were labeled as described previously (Flaming et al., 1993) using MAbs obtained from VMRD,
Inc. (Pullman, WA). T cell subsets were identified using antibodies reactive with
CD2 (CH128A), CD4 (CACTI 38A), or CDS (CACT80C). In addition, a MAb to WC1 90
(BAQ4A) was used to identify the large proportion of circulating gamma-delta T cells that express this marker in cattle. A MAb against bovine IgM (Blg73A) was used to identify B cells, while monocytes were stained with GM1 (DH59B). A mouse MAb against irrelevant antigen served as an isotype control.
After incubating 10® cells with 50 nl of primary antibodies (diluted 1:100), cells were washed and stained with 50 nl of a 1:100 dilution of fluorescein isoththiocyanate-conjugated goat anti-mouse Ig (heavy and light chain) from
CalTag Laboratories Inc. (South San Francisco, CA). Washed cells were fixed in
2% parafomnaldehyde in PBS. Fluorescence was determined at the Iowa State
University Cell and Hybridoma Facility using a Coulter XL-MCL flow cytometer equipped with standard software. Between 5 x 10^ and 1 x 10^ cells were analyzed for each sample. Live cells were discriminated on the basis of their forward and side scatter parameters.
Vaccinations
At 34 months p.i., all cattle were immunized intramusculariy with a commercially available tetanus toxoid (TT) vaccine (Tetanus Toxoid, Fort Dodge
Laboratories, Fort Dodge, lA). A second dose of TT vaccine was given to all cattle
10 months later. The BLV-only and control cattle were later vaccinated twice with bovine herpes virus-1 (BHV-1). The first immunization contained attenuated live
BHV-1 (Sanofi, Overiand Park, KS), and was given at 58 months p.i. In order to boost humoral responses, both groups were vaccinated two months later with a killed, gene-deleted BHV-1 vaccine (IBR/Marker, SyntroVet, Lenexa, KS). 91
Serology
An ELISA to detect TT-specific antibodies in cattle was perfomned as
previously described (Roth et al., 1984), with some modifications. Non-adjuvanted
TT antigen for use in the ELISA was obtained from Fort Dodge Laboratories (Fort
Dodge, lA). Immulon 4 plates (Costar, Cambridge, MA) were coated at 4°C
overnight with 1.8 iig TT per well diluted in 100 |xl antigen binding buffer (0.02 M
sodium carbonate/bicariDonate, pH 9.6). Plates were washed with phosphate
buffered saline (PBS) in an automatic plate washer, and wells were blocked by
adding 220 jil blocking solution (2% gelatin in PBS, pH 9.5-9.7) per well and
incubating for 20 minutes at 39°C. Beginning with a 1:8 dilution for each sample, twofold serial dilutions were made across each row, for a final volume of 100 p.l/well. Each plate contained a row of negative control wells, which contained
similar dilutions of pooled sera from several cattle lacking detectable anti-TT titers.
Serum samples known to contain high levels of anti-TT antibodies were also included as positive controls. Plates were covered and incubated at 39°C for 20 minutes. After washing, 100 jxl of a 1:2000 dilution of peroxidase-conjugated anti- bovine IgG heavy and light chain antibody (Kirkegaard and Perry, Gaithersburg,
MD), was added to each well. Plates were incubated at 39° for 20 minutes, followed by washing and addition of 100 p.1 freshly mixed ABTS peroxidase substrate (Kirkegaard and Perry). Plates were incubated in the dark at 25°C for 10 minutes, and 100 jil of 1 % SDS was added to stabilize color. Optical densities
were immediately read at 405 nm. Means and standard deviations were computed
for each dilution of test samples and negative control wells for all plates tested that 92 day. Anti-TT titer was defined as the highest dilution of each serum sample that gave absorbance readings greater than 3 standard deviations above the mean of all negative wells for that dilution.
The control and BLV-infected groups of cattle later received two BHV-1 vaccinations. Semm samples were obtained at various inten/als and stored at
-20°C prior to submission to the Iowa State University Veterinary Diagnostic
Laboratory for determination of BHV-1 serum neutralizing antibody titers. Endpoint neutralization titers were defined as the highest serum dilution that resulted In
100% inhibition of BHV-1 cytopathic effects in vitro (Jenney and Snyder, 1981).
Isolation of PBMC
To isolate PBMC for use in lymphocyte proliferation assays, approximately
200 ml of peripheral blood was collected from each animal into acid citrate dextrose, centrifuged, and buffy coats were harvested as previously described
(Roth et al., 1982). Contaminating red blood cells were lysed with two volumes of cold distilled water, followed by restoration of nonnal osmolality with triple-strength
PBS. Cells were washed twice in sterile PBS and cell numbers determined with an electronic particle counter. This procedure consistently yielded greater than 90% mononuclear cells as determined by Wright-Giemsa staining, and were 90-95% viable by Trypan blue exclusion staining. 93
Lymphocyte proliferation assays
Prior to placing In cell culture, PBMC were diluted to 1 x 10® cells/ml In RPMI
1640 (GIBCO Laboratories, Chagrin Falls, OH) with 10% fetal calf serum, 100 U ml-i penicillin, and 100 |i.g mh"" streptomycin. Cells were then added to 96 well flat- bottomed tissue culture plates at a density of 2 x 10^ cells/well. Triplicate wells were either left untreated or were stimulated with final concentrations of 7.5 ng/ml concanavalin A (Con A) (Sigma), 0.5 |j.g/ml non-ad]uvanted TT, or 25 |xl/ml (1.25 p.g/ml) of heat-Inactivated BHV-1. Cells were Incubated at SQ'C with 5% CO2 for either two (Con A) or four (TT or BHV-1) days, then labeled with 0.075 ^iCi/well ^H- thymldine for 16-18 hours. Cells were harvested onto glass filter mats with a PHD
Cell Harvester (Cambridge Technology, Inc., Watertown, MA), placed in vials with 5 ml scintillation fluid, and ^H-thymidine uptake was determined with a MInaxI Tri-
Carb 400 Series scintillation counter (Packard Instruments, Downer's Grove, IL).
Differences in counts per minute (DCPM) were calculated by subtracting average
CPM in unstimulated wells from average CPM In treated wells for each animal, and stimulation indexes (SI) were calculated by dividing average treated CPM by average unstimulated CPM.
Mixed leukocyte reactions
Stimulator cells for one-way mixed leukocyte reactions consisted of PBMC
Isolated from each of the three control animals, as well as from three other cattle that tested negative for antibodies to BSV, BIV, or BLV. Thus, lymphocyte 94 proliferative responses from each of the control animals were tested against PBMC from five animals, while all other cattle were tested for reactivity to six animals.
Stimulator cells were inactivated by incubation of 1 x 10^ cells/ml with 50 jig/ml mitomycin C (Sigma) in the dark for 30 minutes at 39°C, followed by three washes in sterile PBS. Round-bottomed tissue culture plates were prepared with 2 x 10^
PBMC in 100 jxl/well from each animal to be tested. Either 2x105or2x104 stimulator cells from each of the six non-infected animals were then added to triplicate wells for each test animal, for a total final volume of 200 ni/well. Additional wells received stimulator cells alone to insure that mitomycin C treatment prevented cell proliferation. Plates were incubated at 39°C in 5% CO2 for 6 days, then cells were labeled with 3H-thymidine and harvested as described above.
Mean responses against stimulator cells from all (five or six) animals were analyzed.
Statistical analysis
For comparisons between ail four groups, data were analyzed as a completely randomized design with a 2 X 2 factorial an*angement of treatments.
Main effects of BIV and BLV infection as well as viral interaction were examined. A main effect comparison tests the effect of an individual virus by comparing all animals that have been infected (for BIV it would be the BIV and BIV/BLV groups) versus all animals not infected with that particular virus (BLV and control groups).
This type of design is more powerful and efficient than simply comparing singly infected animals (BIV) with the control group. A significant viral Interaction would 95 indicate that the effects of one (or both) viruses was not the same in animals that were infected with that virus alone versus animals that were infected with both viruses. Lymphocyte proliferation data for all four groups were analyzed using a repeated measures analysis. Titer data and TT-specific lymphocyte proliferation data were analyzed by day using an ANOVA to determine main effects and interaction of the factorial. All titer data were transformed (log 2) prior to analysis.
In the studies involving vaccination with BHV-1, only control and BLV-only groups were included. These data were analyzed using a iog2 transformation of the raw titers and comparing group means by using the T test. Lymphocyte proliferation data were analyzed by day post-vaccination using the T test.
Results
Confirmation of viral infection
As reported previously, BIV had been isolated from the PBMC of all BIV- infected calves by two weeks p.i. (Carpenter et al., 1992, Isaacson et al., 1995), and
BLV was isolated from the PBMC of all BLV-inoculated calves by eight weeks p.i.
(Alexandersen et al., 1993, Isaacson et al., 1995). One animal in the BlV-only group also tested positive for BSV, but remained in the study. In addition, all BIV- inoculated cattle seroconverted to BIV core antigens by four weeks p.i., and all had persistent antibody reactivity to a recombinant fusion protein containing the amino terminal portion of the BIV transmembrane envelope protein gp42 throughout the present study (Isaacson et al., 1995). All BLV-inoculated animals tested positive for 96 antibodies to BLV gp51 by an agar gel immunodiffusion test prior to beginning the present study (data not shown).
Clinical monitoring
No overt signs of illness were obsen^ed in any of the cattle during the study period. However, one animal in the BlV-only group (#346) died of an atypical non-
BLV lymphosarcoma at seven months p.i. (Rovid et al., 1996), and therefore was not a part of the present study.
Peripheral blood counts
Analysis of total and differential WBC and lymphocyte subsets in these cattle through the fourth year post-inoculation (p.i.) have been reported in a separate publication (Flaming et al, in press), and data from that study are repeated here only where relevant to the present study. Cattle infected with BIV had no long-term, consistent changes in either total or differential WBC counts, or in percentages of lymphocytes staining for B or T cell markers through 49 months p.i. (Flaming et al., in press), or in the present study (Table 1). In contrast, BLV-infected cattle had increased numbers of total WBC, mononuclear cells, and cells expressing CD2,
CD4, WC1, or IgM in the fifth year p.i. (Table 1). Similar increases in these lymphocyte subsets have been previously reported in these BLV-infected cattle, although the alterations were not always statistically significant in each of the first four years p.i. (Flaming et al., in press). Neither BIV nor BLV infection had any Table 1. Comparison^ of cell numbers^ in peripheral blood for the fifth year p.i.
Group WBC Granulocytes^ Mononuclears^ CD2+ CD4+ CD8+ WC1+ lgM+ CD4:CD8 CD2:lgM
Control 5730 2530 3210 1580 950 490 280 540 1.95 2.99 (±S.E.) ±603 ±70 ±530 ±350 ±190 ± 120 ± 120 ±340 ±11 ±1.41
BIV 5267 2140 3130 1530 860 550 280 730 1.57 2.13 (±S.E.) ±473 ±569 ±980 ±390 ± 170 ± 140 ±90 ±230 ± 09 ±2
BIV/BLV 7880L"' 2190 5690L"* 2150L* 1410L"* 730 780'-** 1510L*' 2.22 1.40 (±S.E.) ± 1708 ±437 ± 1590 ±520 ±330 ±370 ±443 ±664 ±•78 ±.52
BLV 6620L''' 2157 4460L*" 1900L' 1590L*** 730 720L- 890L" 2.23 2.10 (±S.E.) ±545 ±617 ±720 ±516 ±445 ±241 ±341 ±329 ± .38 ±.75
a L = BLV effect, I = BIV effect. X = interaction; * = p < 0.1, " = p < 0,05, "' = p < 0.01) cells per /7l of blood ^ neutrophils, eosinophils, and basophils ^ monocytes and lymphocytes 98 significant effect on CD4+: CD8+ orT cell: B cell ratios, (Table 1) Of the ten BLV- infected cattle, only one animal (in the BIV/BLV group) had high enough lymphocyte counts to be considered persistently lymphocytotic through the first four years p.i. (Flaming et al., in press). In the fifth year p.i., the lymphocyte counts of this animal were still slightly above the upper prediction limits developed by
Thumnond et al. (1990), while two others (one BIV/BLV, one BLV only) were considered borderline (data not shown).
Humoral responses to tetanus toxoid
Antibodies specific for TT were quantitated by ELISA before and after TT vaccinations. Tetanus antibody titers were either undetectable or less than 1:8 in all animals prior to vaccination. Following primary vaccination, anti-TT titers were significantly lower in the BlV-infected group at day 33, and remained lower, though not significantly so, through day 95 (Fig. 1(a)). After the second TT vaccination, the
BIV/BLV and BLV groups had approximately fourfold higher humoral responses to the recall antigen, beginning at about 2 weeks post-vaccination (Fig. 1(b)).
Therefore, coinfection with BIV appeared to have no effect on this enhancement of humoral responsiveness by BLV. The BLV effect was significant at the p < 0.1 level for seven of eight samplings taken between 22-135 days following the second vaccination (p < 0.01 for 5 samplings). 99
2000 1500- A— Control BIV C •— BIV/BLV (DCO 1000- BLV (D E 500 o oCD 10 20 30 40 50 60 70 Days Post-1 st TT Vaccination B. 35000 - (U 30000 - 25000• cz < 20000 - —A— Control c BIV a>to BIV/BLV 15000- •— BLV o "S 10000- E oO) O 5000- 20 40 60 80 100 120 140 Days post-2nd TT vaccination Figure 1. Effect of experimental BIV and/or BLV infection on humoral responses to tetanus toxoid (TT). Anti-TT titers were determined by ELISA. Geometric mean titers were calculated for statistical analysis, and these means were converted back to titers for plotting group means. Statistical differences are indicated as follows: I = BIV effect; L = BLV effect; X = interaction between the two vimses (* = p < 0.1; ** = p< 0.05; *** = p< 0.01). A. Primary responses. B. Secondary responses. 100 Lymphocyte proliferation Lymphocyte proliferative responses were determined by 3H-thymldine uptake in PBMC cultured either without additional stimulation (spontaneous lymphocyte proliferation or SLP), or following exposure to Con A, antigens, or allogeneic MHC. Group means for both stimulation indexes (SI) and differences in counts per minute (DCPM) were calculated and analyzed, but only DCPM results are depicted in Figure 2. Levels of SLP were significantly higher in the BLV- infected cattle (Fig. 2(a)). Although there was considerable variation in SLP within the BLV groups (ranges were typically 100-1000 CPM for controls, 100-20,000 CPM for BLV-infected cattle), SLP values remained relatively consistent for each individual animal during the study (data not shown). As shown in Rgure 2(a), BIV Infection was associated with a significant (p < 0.05) decrease in levels of ^H-thymidine incorporation by lymphocytes stimulated with the T cell mitogen Con A in the fifth year p.i.. A previous study of these cattle had shown that lymphocyte proliferation in response to Con A during the fourth year p.i. tended to be decreased in BlV-infected cattle, but not in cattle infected with both BIV and BLV (Flaming et al., in press). A similar effect was seen in the present study, although the interaction between the two viral infections was not statistically significant. In addition, lymphocytes from BLV-infected animals had slightly enhanced responses to Con A in the fifth year p.i. (Fig. 2(a)). A similar response to Con A was reported for these BLV-infected cattle in the third year p.i., but in the fourth year p.i., lymphocyte proliferation in response to Con A was slightly lower in the BLV-infected groups compared to uninfected controls (Flaming et al, in press). Figure 2. Effect of bovine retroviral infections on lymphocyte proliferation in vitro. A. Spontaneous lymphocyte proliferation (SLP) refers to the counts per minute (CPM) due to spontaneous uptake of 3H-thymidine by PBMC in wells receiving no exogenous stimulation. Lymphocyte responses to Con A and mixed leukocyte reactions (MLR) are reported as differences in counts per minute (DCPM). Since the SLP and Con A treatments were included in all experiments as controls, these data represent means from 6-8 experiments conducted between 44 and 64 months p.i. Data were analyzed by repeated measures analysis. MLR were performed twice for each stimulator; responder cell ratio. A ratio of 1 ;1 indicates equal numbers of stimulator and responder cells were added to each well; in the 1 ;10 experiments, only one stimulator cell was added for every 10 responder cells. B. TT-specific lymphocyte proliferation by PBMC stimulated with TT in vitro. Viral effects and P values are represented as described for Rg. 1. 102 50000- 40000- a Control 30000 • BIV • BIV/BLV • BLV 20000- 10000- SLP ConA MLR 1:1 MLR 1:10 Treatment 14000- 12000 10000 8000 • Control • BIV ZI BIV/BLV 6000 B BLV 4000 2000 -35 7 29 63 91 Days Post-2nd TT Immunization 103 Thus, lymphocyte responses to Con A generally were not dramatically affected when DCPM were analyzed. However, when SI were compared instead, Con A induced lymphocyte proliferation appeared to be significantly decreased in the BLV-infected groups due to the higher levels of SLP (data not shown). Exposure of PBMC to allogeneic MHC in the fonn of mitomycin-C-inactivated PBMC from six uninfected cattle resulted in enhancement of ^H-thymidine uptake compared to unstimulated cells, but no significant differences in DCPM between the groups (Fig. 2(a)). As with Con A, when SI were compared, BLV infection appeared to cause a significant decrease in responses to allogeneic MHC, (data not shown), but this was entirely attributable to the increased background counts in PBMC from BLV-infected cattle. Because of the observation that bovine retroviral infections can affect humoral responses to tetanus toxoid vaccination in vivo, lymphocyte proliferation was also assessed in cultured PBMC stimulated with non-adjuvanted TT antigen. As shown in Figure 2(b), BIV or BLV Infection status had no consistent effect on the uptake of ^H-thymidine by PBMC exposed to TT in vitro. There was considerable variation in responses between different sampling dates, as well as between different animals within each group. Lymphocyte proliferation in response to TT appeared to be Increased in the BLV-only group 63 days post-vaccination, but this was predominantly due to exceptionally high values for one animal on that sampling date, and thus the means were not significantly different at the p < 0.05 level. 104 Effect of BLV infection on immune responses to BHV-1 In order to assess whether the BLV-induced enhancement of humoral responses was unique to tetanus toxoid, the five cattle infected with BLV and three uninfected controls were vaccinated twice with BHV1. As shown in Rg. 3(a), the BLV infected cattle had increased antibody responses to BHV-1, especially after the second vaccine was administered. However, geometric mean titers were significantly different between the groups at the p < 0.05 level for only two sampling dates. As shown in Fig. 3(b), vaccination with attenuated BHV1 increased the uptake of ^H-thymidine by lymphocytes from cattle treated with BHV1 in vitro. However, there were no consistent differences in BHV1-stimulated lymphocyte proliferation in PBMC cultures from BLV-infected and control cattle. Discussion In this study various immune parameters were examined in cattle between three and five years after experimental inoculation with BIV and/or BLV. Considering observations from other lentiviral infections, we hypothesized that a decline In antigen-specific immune responses might occur relatively early in the course of BIV infection, prior to any potential changes in circulating lymphocyte numbers or development of clinical signs. However, BIV infection caused no consistent alterations in humoral responses to vaccination with tetanus toxoid (TT), 105 400- 300- 200 S 100- Vacc. #1 vacc. #2 Day 20000 15000- • Control 10000- B BLV 5000- Vacc. #1 Vacc. #2 Day Figure 3. Effect of BLV infection on humoral and lymphocyte proliferative responses to BHV-1 vaccination. A. Serum neutralizing responses to BHV-1 vaccinations In control and BLV-only groups. Cattle were immunized first with modifed live BHV-1 (Vacc. #1), then with a killed BHV-1 product (Vacc. #2). B. Lymphocyte proliferation in PBMC stimulated with BHV-1 in vitro. Viral effects and P values are represented as described for Figure 1. 106 or in lymphocyte proliferative responses to TT or allogeneic MHC. Lymphocyte proliferative responses to Con A were decreased in BlV-infected cattle during the fifth year p.i. compared to uninfected controls. The mechanism of BlV-induced suppression of lymphocyte responsiveness to this T cell mitogen is unknown, but it may differ from that seen in HIV infection, where decreased lymphocyte responses to mitogens usually occur after the decline in responses to antigens and allogeneic MHC, and often coincides with the onset of clinical AIDS (Shearer et al., 1986). The results reported here suggest that experimental infection of cattle with BIV does not result in any obvious deficits in acquired immune responses duringthe first five years p.i. It is important to note, however, that similar to most previous studies of the effects of BIV infection, we examined cattle inoculated with the original R-29 strain of BIV, which may have been attenuated due to repeated passage in cell culture. Lower-passage BIV isolates are now available (Suarez et al, 1993), but there have been no reports thus far of the immune status of cattle infected for more than four months with these isolates. A recent study suggested that BIV may have played a role in the chronic health problems of a herd with a high rate of BIV infection (Snider et al., 1996). Because it is difficult to discern the role of BIV in natural infections due to the possible influence of other infectious agents or factors such as poor nutrition, toxins, and stress. It will be important to continue to investigate the effects of long-term experimental BIV infection on bovine immune functions under various experimental conditions. In order to examine the potential role of BLV as a co-factor in BIV infection, two groups of cattle in the present study were inoculated with BLV. Immune parameters of cattle infected with both BIV and BLV did not differ significantly from 107 those Infected with BLV only, indicating a lack of interaction between these viruses under these experimental conditions. In contrast, both groups of BLV-lnfected cattle had significantly Increased secondary antibody responses to TT vaccination compared to controls, and the BLV-lnfected group had consistently higher serum virus neutralizing antibody titers following vaccination with BHV-1. Cattle infected with BLV also had elevated numbers of circulating mononuclear cells, which was attributable to increases In both B and T cells, including WC1-expressing gamma- delta T cells. Lymphocytes from BLV-infected cattle exhibited spontaneous lymphocyte proliferation and marginally increased proliferative responses to Con A (as determined by DCPM), but there was no consistent BLV effect on lymphocyte proliferative responses to TT, BHV-1, or allogeneic MHC. There have been few previous studies of the effect of BLV infection on Immune responses to non-BLV antigens. Cattle that develop BLV-induced lymphoma often exhibit signs of Immunosuppression In the terminal stages of disease (Jacobs et al., 1981, Yamamoto et al., 1984), but data on immune responses of clinically normal, BLV-infected cattle with or without persistent lymphocytosis are less clear. It has been known for many years that lymphocytes from some BLV-infected cattle undergo increased spontaneous proliferation in vitro. Decreased lymphocyte proliferative responses to mitogens, when calculated as stimulation Indexes, have sometimes been Interpreted as evidence for BLV- induced immunosuppression (Muscoplat et al., 1974b, Takashima and Olson, 1980), yet no such effect was observed when differences In CPM were analyzed Instead (Williams et al, 1988b). Similarly, because of the increased spontaneous proliferation of lymphocytes from BLV-infected cattle In the present study. 108 consideration of stimulation indexes only would have led to the conclusion that lymphocyte responses to Con A, allogeneic MHC, TT, and BHV-1 were significantly inhibited In the BLV-infected groups. However, our obsen^ation that BLV-infected cattle have increased secondary antibody responses to two different antigens, measured either as total antibody to TT or as BHV-1-neutralizing activity, suggests that BLV infection may enhance secondary antibody responses under a fairly broad range of immunization conditions. About 30% of BLV-infected cattle develop a persistent lymphocytosis that is often characterized by greatly increased numbers of circulating B cells (Muscoplat et al., 1974a, Takashima et al., 1977, Williams et al., 1988a, Gatei et al., 1989, Taylor et al., 1992). Several of these reports also described increased B cell numbers in BLV infected, non-lymphocytotic cattle. Although only one animal In the present study could be classified as persistently lymphocytotic, cattle in both BLV-infected groups had generally increased numbers of circulating B ceils. It seems reasonable to assume that an increase in numbers of antibody-producing cells would tend to enhance humoral responses to vaccination. What remains unclear, however, is whether the modest increase in B ceil numbers seen in these BLV-infected cattle completely accounts for the fairly dramatic increase in antibody responses reported here. There have been some conflicting reports on the effect of BLV infection on numbers of circulating T cells. In BLV-infected, lymphocytotic cattle, T cell numbers have been reported to be increased (Williams et al., 1988a, Stone et al., 1994), unaffected (Taylor et al., 1992), or decreased (Gatei et al., 1989, Sordillo et al., 1994). Similariy, increased (Williams et al., 1988a), unchanged (Gatei et al, 1989), 109 and decreased (Sordillo et al., 1994) T cell numbers have been reported in BLV- infected non-lymphocytotic cattle. The factors that determine T cell numbers in BLV infection are unknown, but it is likely that the increase In CD4+ T cell numbers seen in the present study contributed to the enhancement of antibody responses. These T cells would be expected to provide increased amounts of costimulatlon and/or soluble cytokines capable of enhancing B cell proliferation and differentiation. Moreover, increased numbers of B cells from BLV-infected lymphocytotic cattle were recently reported to express the Inducible alpha chain of the IL-2 receptor, suggesting that they might proliferate more In response to this cytokine (Stone et al., 1994). Increased amounts of IL-2 have also been detected In supematants of mitogen-stlmulated lymphocytes from BLV-infected, lymphocytotic cattle (Sordillo et al., 1994). Taken together with our In vivo results, these studies suggest that enhanced T cell-B cell interactions in BLV-infected cattle may result in increased antibody responses to T cell dependent antigens. Because of the potential relevance of BLV-induced Immune activation to the understanding of BLV pathogenesis. It will be important to determine whether enhancement of antibody responses occurs in cattle with BLV infections of varying durations, as well as In naturally acquired Infections. In addition, the knowledge that BLV infection may enhance antibody responses in cattle provides a strong rationale for screening cattle for BLV infection prior to Inclusion in vaccine trials or other immunological studies. BLV Infection has been associated with decreases In milk fat percentage and milk production (Wu et al., 1989, Da et al., 1993), as well as Increased culling rates (Brenner et al., 1989). An Interesting question to be answered Is whether stress due to chronic BLV-induced immune activation may 110 contribute to these BLV effects on production. Overall, much remains to be learned about the long-term effects of BLV and BIV infection on the host immune system. Acknowledgements The authors wish to thank Tom Skadow and Andrea Dom for excellent technical assistance. Agar gel immunodiffusion tests for BLV antibodies were perfomned by Dennis Orcut. This work was supported by PHS grants R01 CA59125 and AI07378. References BeiIke, M.A., Greenspan, D.L., Impey, A., Thompson, J., and Didier, P.J., 1994. Laboratory study of HIV-1 and HTLV-I/II coinfection. J. Med. Virol. 44: 132-143. Bishop, S.A., Williams, N.A., GruffyDD-Jones, T.J., Harbour, D.A., and Stokes, C.R., 1992. An early defect in primary and secondary T cell response in asymptomatic cats during acute feline immunodeficiency virus (FIV) infection. Clin. Exp. Immunol. 90: 491-496. Brenner, J., Van-Haam, M., Savir, D., and Trainin, Z., 1989. The implication of BLV infection in the productivity, reproductive capacity and survival rate of a dairy cow. Vet. Immunol. Immunopathol. 22: 299-305. Brownlie, J., Collins, M.E., and Heaton, P., 1994. Bovine immunodeficiency-like virus~a potential cause of disease In cattle? Vet. Record 134: 289-291. Bumy, A., Cleuter, Y., Kettmann, R., Mammerickx, M., Marbaix, G., Portetelle, D., Van Den Broeke, A., Willems, L., and Thomas, R., 1988. Bovine leukaemia: facts and hypotheses derived from the study of an infectious cancer. Vet. Microbiol. 17: 197-218. Carpenter, S., Miller, L.D., Alexandersen, S., Whetstone, C.A., Van Der Maaten, M.J., Viuff, B., Wannemuehler, Y., Miller, J.M., and Roth, J.A., 1992. Characterization of eariy pathogenic effects after experimental infection of calves with bovine immunodeficiency-like virus. J. Virol. 66: 1074-1083. 111 Da., Y., Shanks, R.D., Stewart, J.A., and Lewin, H.A., 1993. Milk and fat yields decline in bovine leukemia virus-infected Holstein cattle with persistent lymphocytosis. Proc. Natl. Acad. Sci. 90: 6538-6541. Flaming, K., Van Der Maaten, M,, Whetstone, C., Carpenter, S., Frank, D., and Roth, J., 1993. The effect of bovine immunodeficiency-like virus infection on immune function in experimentally infected cattle. Vet. Immunol. Immunopathol. 36: 91-105. Flaming, K.P., Frank, D.E., Carpenter, S., and Roth, J.A. Longitudinal studies of immune function in cattle experimentally infected with bovine immunodeficiency like virus and/or bovine leukemia virus. Vet. Immunol. Immunopathol., in press. Gatei, M.H., Brandon, R.B, Naif, H.M., McLennan, M.W., Daniel, R.C.W., and Lavin, M.F,, 1989. Changes in B cell and T cell subsets In bovine leukaemia virus-infected cattle. Vet. Immunol. Immunopathol. 23: 139-147. Gonda, M.T., Oberste, M.S., Garvey, K.J., Pallansch, L.A., Battles, J.K., Pifat, D.Y., and Nagashima, K., 1990. Contemporary developments in the biology of the bovine immunodeficiency-like virus. In: H. Schellenkens and M.C. Horzinek (Editors), Animal Models in AIDS, Elsevier Science Publishers B. V, pp. 233-255. Isaacson, J.A., Roth, J.A., Wood, C., and Carpenter, S., 1995. Loss of gag-specific antibody reactivity in cattle experimentally infected with bovine immunodeficiency like virus. Viral Immunology 8: 27-36. Jacobs, R.M., Valli, V.E.O., and Wilkie, B.N., 1981. Response of cows with lymphoma to the intradennal injection of tumor cell antigens and phytohemagglutinin. Can. J. Comp. Med. 45: 43-50. Jenney, E.W., and Snyder, M.L., 1981. Infectious bovine rhinotracheitis neutralization test. In: Serologic microtitration techniques, USDA/APHIS, National Veterinary Service Laboratory, Ames, lA, p. 35-36. Martin, S.J., O'Neill, T.P., Bilello, J.A., and Eiseman, J.L., 1991. Lymphocyte transformation abnormalities in bovine immunodeficiency-like virus infected calves. Immunol. Lett, 27: 81-84. Muscoplat, C.C., Johnson, D.W., Pomeroy, K.A., Olson, J.M., Larson, V.L., Stevens, J.B., and Sorenson, D.K., 1974a. Lymphocyte surface immunoglobulin: Frequency in normal and lymphocytic cattle. Am. J. Vet. Res. 35: 593-595. Muscoplat, C.C., Aihaji, I., Johnson, D.W., Pomeroy, K.A., Olson, J.M., Larson, V., Stevens, J.B., and Sorenson, D.K., 1974b. Characteristics of lymphocyte responses to phytomitogens: Comparison of responses of lymphocytes from nomial and lymphocytotic cows. Am. J. Vet. Res. 35:1053-1055. 112 Onuma, M., Koomoto, E., Fumyama, H., Yasutomi, Y., Taniyama, H., Iwai, H., and KawakamI, Y., 1992. Infection smd dysfunction of monocytes Induced by experimental Inoculation of calves with bovine Immunodeficiency-like virus. JAIDS 5: 1009-1015. Pinching, A.J., 1991. Antibody responses in HIV infection. Clin. Exp. Immunol. 84: 181-184. Roth, J.A., Kaeberie, M.L. and Griffith, R.W., 1981. Effects of bovine viral diarrhea vims Infection on bovine polymorphonuclear leukocyte function. Am. J. Vet. Res. 41: 244-250. Roth, J.A., Kaeberie, M.L, and Hsu, W.H., 1982. Effects of ACTH administration on bovine polymorphonuclear leukocyte function and lympocyte blastogenesis. Am. J. Vet. Res. 43:412-416. Roth, J.A., Kaeberie, M.L., and Hubbard, R.D., 1984. Attempts to use thiabendazole to Improve the Immune response in dexamethasone-treated or stressed cattle. Immunophamiacology 8: 121-128. Rovid, A.H., Carpenter, S., and Roth, J.A., 1995. Monocyte function in cattle experimentally Infected with bovine immunodeficiency-like virus. Vet. Immunol. Immunopathol. 45: 31-45. Rovid, A.H., Carpenter, S., Miller, L.D., Flaming, K,P, Long, M.J., Van Der Maaten, M.J., Frank, D.E., and Roth, J.A., 1996. An atypical T-cell lymphosarcoma in a calf with bovine Immunodeficiency-like virus infection. Vet. Pathol. 33: 457-459. Sagata, N., Yasunaga, T., Tsuzuku-Kawamura, J., Ohishi, K., and Ogawa, Y., 1985. Complete nucleotide sequence of the genome of bovine leukemia virus: Its evolutionalry relationship to other viruses. Proc. Natl. Acad. Sci. USA 82: 677-681. Shearer, G.M., Bernstein, D.C., Tung, K.S.K., Via, C.S., Redfield, R., Salahuddin, S.Z, and Gallo, R.C., 1986. A model for the selective loss of major histocompatibility complex self-restricted T cell immune responses during the development of acquired immune deficiency syndrome (AIDS). J. Immunol. 137: 2514-2521. Snider, T.G., Luther, D.G., Jenny, B.F., Hoyt, P.G., Battles, J.K., Ennis, W.H., Balady, J., Blas-Machado, U., Lemarchand, T.X., and Gonda, M.A., 1996. Encephalitis, lymphoid tissue depletion and secondary diseases associated with bovine immunodeficiency virus in a dairy herd. Comp. Immunol. Microbiol. Infect. Dis. 19: 117-131. Sordillo, L.M., Hicks, C.R., and PighettI, G.M., 1994. Altered interieukin-2 production by lymphcyte populations from bovine leukemia virus-infected cattle. Proc. Soc. Exp. Biol. Med. 207: 267-273. 113 Stone, D.M., McElwain, T.F., and Davis, W.C. 1994. Enhanced B-lymphocyte expression of IL-2Ra associated with T lymphocytosis in BLV-infected persistently lymphocytotic cows. Leukemia 8:1057-1061. Suarez, D.L, Van der Maaten, M.J., Wood, C. and Whetstone, C.A., 1993. Isolation and characterization of new wild-type isolates of bovine lentivirus. J. Virol. 67: 5051-5055. Takashima, I., and Olson, C., 1980. Effect of mitogens and anti-bovine leukosis virus serums on DNA synthesis on lymphocytes from cattle. Europ. J. Cancer 16: 639-645. Taylor, B.C., Stott, J.L, Thurmond, M.A., and Picanso, J.P., 1992. Alteration in lymphocyte subpopulations in bovine leukemia virus-infected cattle. Vet. Immunol. Immunopathol. 31: 35-47. Thumnond, M.C., Carter, R.L, Picanso, J.P., and Stralka, K., 1990. Upper-normal prediction limits of lymphocyte counts for cattle not infected with bovine leukemia virus. Am. J. Vet. Res. 51:466-470. Torten, M., Franchini, M., Barlough, J.E., George, J.W., Mozes, E., Lutz, H., and Pedersen, N.C., 1991. Progressive immune dysfunction in cats experimentally infected with feline immunodeficiency virus. J. Virol. 65: 2225-2230. Van der Maaten, M.J., Boothe, A.D., and Seger, C.L., 1972. Isolation of a virus from cattle with persistent lymphcytosis. J. Natl. Cancer. Inst. 49:1649-1657. Williams, D.L, Amborski, G.F., and Davis, W.C., 1988. Enumeration of T and B lymphocytes in bovine leukemia virus-infected cattle, using monclonal antibodies. Amer. J. Vet. Res. 7:1098-1103. Williams, D.L, Barta, O., and Amborski, G.F., 1988(b). Molecular studies of T lymphocytes from cattle infected with bovine leukemia virus. Vet Immunol. Immunopathol.19: 307-323. Wu, M.C,, Shanks, R.D., and Lewin, H.A., 1989. Milk and fat production in dairy cattle influenced by advanced subclinical bovine leukemia virus infection. Proc. Natl. Acad. Sci. 86: 993-996. Yamamoto, S,, Onuma, M., Kodama, H., Mikami, T., and Izawa, H., 1984. Suppression of natural cytotoxic activity of lymphocytes from cattle and sheep during the progress of bovine leukosis. Vet. Micro. 9:105-111. 114 CHAPTER 5. INCREASED MHC CLASS II AND CD25 EXPRESSION ON LYMPHOCYTES FROM NON-LYMPHOCYTOTIC CATTLE EXPERIMENTALLY INFECTED WITH BOVINE LEUKEMIA VIRUS A paper to be submitted to Veterinary Immunology and Immunopathology Jeffrey A. Isaacson, Kevan P. Flaming, and James A. Roth Abstract We recently observed that a group of cattle experimentally Infected with bovine leukemia virus (BLV) had enhanced antibody responses to recall antigens. None of the cattle in this group were classified as persistently lymphocytotic, but they did have significantly increased numbers of circulating T and B cells. Although the enhancement of humoral reactivity to non-BLV antigens might have been entirely due to the increase in lymphocyte numbers, we hypothesized that BLV-induced alterations in lymphocyte function could also have played a role. In order to investigate potential mechanisms of BLV-induced immune activation, dual- color flow cytometry was used to compare the expression of MHC Class II (MHC II) molecules and the IL-2 receptor alpha chain (CD25) on lymphocyte subsets in freshly isolated and cultured PBMC from these same BLV-infected cattle (n=5) with that of age-matched controls (n=3). Freshly isolated peripheral blood mononuclear cells (PBMC) from BLV-infected cattle were found to contain a significantly increased percentage of B cells that expressed MHC II molecules compared to non-infected controls (P < 0.01). There were no differences between the groups in MHC II expression by T cells in unstimulated PBMC, but an Increased percentage of CD4+ T cells from BLV-infected cattle expressed MHC II after 20 hours of Con A stimulation (p < 0.05), and MHC II expression was increased on both CD4-t- 115 (p < 0.01) and CD8+ (p < 0.05) T cells from BLV-infected cattle after 68 hours in vitro, even in the absence of exogenous mitogen. Although CD25 expression was not enhanced on freshly Isolated lymphocytes from BLV-lnfected cattle, an increased percentage of B cells from BLV-infected cattle expressed CD25 in 20 hour cultures in the presence (p < 0.05) or absence (p < 0.01) of Con A. Thus, in addition to causing alterations In absolute numbers of circulating lymphocytes, BLV Infection can cause a functional activation of both B and T cells, even in cattle that are non-lymphocytotic. It is likely that these BLV-lnduced alterations in lymphocyte activation status contributed to the previously obsen/ed enhancement of antibody responses In vivo. Abbreviations BLV, bovine leukemia virus; Con A, concanavalin A; HTLV, human T cell leukemia virus; MHC II, major histocompatibility complex class II; PBMC, peripheral blood mononuclear cells; PBS, phosphate-buffered saline; PL, persistent lymphocytosis; sIgM, cell surface IgM. Introduction Bovine leukemia virus (BLV) is an exogenous retrovirus of cattle that predominantly infects B cells (Paul et al., 1977), although BLV provlral DNA has also been detected In T cells, monocytes, and granulocytes (Williams et al., 1988b, Schwartz et al., 1994). Approximately 30 percent of infected cattle eventually develop persistent lymphocytosis (PL), which Is characterized by dramatically increased numbers of circulating slgM+ B cells (Williams et al., 1988a, Gatel et al., 116 1989, Taylor et al., 1992). Although animals with PL are at a greater risk of developing BLV-induced lymphosarcoma, it remains unclear whether PL is a pre neoplastic state or a lymphoproliferative response to the presence of the virus. Because of the difficulty in detecting BLV proteins or BLV mRNA in freshly isolated lymphocytes or tissues, it has been proposed that BLV is functionally latent in vivo, especially in infected cattle that remain hematologically normal (Kettmann et a!., 1980). More recently, the use of PCR-based techniques has allowed the detection of BLV-specific mRNA in PBMC from cattle at all stages of infection (Lagarias and Radke, 1989, Jensen et al., 1991), suggesting the latency of BLV infection is incomplete. One implication of these findings Is that persistent, low levels of BLV expression in vivo may influence host immune functions, even in BLV-infected, non-lymphocytotic cattle. There have been few studies of the effects of BLV infection on host immune system, particularly in cattle that remain apparently healthy. Although cattle In the terminal stages of BLV-induced lymphoma often have clear signs of immunosuppression (Trainin et al., 1976, Jacobs et al., 1981), there is little evidence to suggest that this is a general feature of BLV infection. In contrast, the findings of some recent studies imply that BLV infected cattle may cause an immune activation. The development of PL in BLV-infected cattle and sheep is also characterized by a dramatic expansion of the CD5+ B cell population (Lettesson et al., 1990, Matheise et al., 1992). Although the function of CDS on bovine B cells is unclear, an increased percentage (>15%) of CD5+ B cells from lymphocytotic cattle were shown to be activated cells with respect to increased DNA content and responsiveness to an IL-2-containing supematant (Maltheise et al., 1992). Stone et al. (1994) showed that an increased percentage of B cells from BLV-infected, lymphocytotic cattle expressed the inducible alpha chain of the IL-2 117 receptor (CD25) after stimulation with mitogens or recombinant IL-2 (Stone et al., 1994, 1995). Supematemts from cultured PBMC of BLV-infected, lymphocytotic cattle were also found to contain Increased amounts of IL-2 (Sordillo et al., 1995). Therefore, BLV infection can potentially affect both B and T cell functions. It has been suggested that enhanced interactions between T and B cells may contribute to the activation and proliferation of B cells characteristic of PL (Stone et al., 1995). However, because most previous studies have focused on BLV-infected cattle with PL, an important question is whether BLV induces similar lymphocyte phenotypic alterations in the majority of BLV-infected cattle which do not develop PL. In an accompanying paper, we reported that BLV infected cattle had enhanced secondary antibody responses to two recall antigens, tetanus toxoid and bovine herpesvirus-1. Although none of the animals singly infected with BLV were considered to be lymphocytotic, BLV Infection induced significant increases (p < 0.05) in numbers of circulating T and B cells. Therefore, it was unclear whether the enhancement of antibody responses to non-BLV antigens was entirely attributable to the increased lymphocyte numbers, or whether functional changes in lymphocyte activity were also a contributing factor. In the current study, freshly isolated and cultured PBMC from the same BLV-infected cattle that had enhanced humoral responses to recall antigens were examined by dual-color flow cytometry for expression of Immune activation markers on specific lymphocyte subsets. The markers used were MHC Class II (MHC II) molecules and the IL-2 receptor alpha chain, CD25. There was evidence of increased activation of both B and T lymphocytes of BLV-infected, non-lymphocytotic cattle compared to uninfected controls. The results of these two studies indicate that a BLV-induced polyclonal lymphocyte activation may have played a role in the enhancement of antibody responses to non-BLV antigens. 118 Materials and Methods Animal inoculations Five colostrum-deprived, castrated male Holstein calves were Inoculated with BLV, and three calves were sham-inoculated as previously described (Carpenter et al., 1992, Isaacson et al, submitted). The cattle had been infected for four and one-half years at the beginning of the present study, which continued over a period of about one year. Peripheral blood cell counts Total and differential white blood cell counts were detemnined on whole blood, and lymphocyte subset percentages were quantitated in freshly isolated PBMC by flow cytometric analysis as previously described (Flaming et a!., 1993, Isaacson et al., submitted). Isolation of PBMC In order to provide sufficient numbers of PBMC for evaluating the effects of different culture conditions on the expression of cell surface mariners, approximately 200 ml venous blood from each animal was collected into acid citrate dextrose, followed by centrifugation, collection of buffy coats, and hypotonic lysis of contaminating erythrocytes (Roth et al, 1982). Cells were then washed 119 twice with sterile PBS and counted. Viability of PBMC prepared In this manner was consistently greater than 95% by Trypan blue dye exclusion. Culture and harvesting of cells For all in vitro studies, PBMC were diluted to 1 x 10® cells ml-"" in RPM11640 (Celox Laboratories, Inc., Hopkins, MN) with 10% fetal calf serum, 100 U ml-i penicillin, and 100 ml""' streptomycin. For each animal, 2 x 10® cells (2 ml) were added to all wells of a 24-well flat-bottomed tissue culture plate, which allowed eight wells per treatment. One set of eight wells served as controls receiving no additional stimulation, while the other wells were treated with either Con A (7.5 |ig ml""*) or PWM (final dilution 1:100). Duplicate sets of plates were set up for each animal, and cells were incubated with 5% CO2 at 39°C for either 20 or 68 hours prior to han/esting. Cells were removed from the wells by first centrifuging plates at 650 x g for 10 minutes, aspirating most of the supematants, then adding 0.2 ml cold trypsin (Difco) In 0.5 mM EDTA at 4'C to loosen adherent cells. After incubating on ice for five minutes, 0.2 ml cold fetal calf serum was added to inactivate the trypsin. Cells were then removed from the wells by vigorous pipetting, and each well was rinsed once with 1 ml cold PBS to remove remaining cells. Staining of wells with crystal violet revealed that very few adherent cells remained in the wells after the washing step. Cells from a given treatment were pooled for each animal, and pooled cells were washed three times with cold PBS-H-, counted, and diluted to 10® viable cells mh"! in PBS++ (PBS with 0.5% BSA and 0.1% sodium azide) for incubation with antibodies. 120 Monoclonal antibodies (MAbs) Mouse MAbs reactive with the following bovine antigens were obtained from VMRD (Pullman, WA): CD4 (CACT83B); CDS (CACT80C); WC1 (BAQ4A); IgM (Blg73A), MHC Class II (TH14B), and the IL-2 receptor alpha chain or CD25 (CACTI 08A). The MAbs against CD4, CDS, WC1, and IgM were all isotype lgG1; those reactive with MCHII and CD25 were lgG2a. Murine lgG1 and lgG2a MAbs of irrelevant specificities sen/ed as isotype controls. For secondary antibodies, affinity-Isolated FITC-conjugated anti-mouse IgGI (M32101) and PE-conjugated anti-mouse lgG2a (M32204) were purchased from Cal Tag Laboratories (San Francisco). Cross-reactivity experiments showed that freshly Isolated PBMC labeled with any of the IgGI primary MAbs showed less than five percent non specific reactivity with the lgG2a-specific reagent, and vice versa (data not shown). Cell staining and flow cytometry For MAb staining of both freshly Isolated and cultured PBMC, 1x10® cells were added to wells of 96 well microtiter plates. Duplicate wells received 25 p.1 of 1:50 dilutions of primary MAb to either CD2, CD4, CDS, IgM, WC1, or GM1. Half of the wells for each treatment then received an additional 25 |il of anti-CD25, and half received anti-MHC II diluted 1:50 (final dilution 1:100 for each MAb), and plates were incubated for 25 to 40 minutes at 4*C. Cells were then washed 3 times in PBS++, and resuspended in 50 ^1 of a mixture of FITC-labeled anti-mouse IgGI and PE-labeled anti-mouse lgG2a (both diluted 1:100). Following another 25 to 40 minutes incubation In the dark at 4* C, cells were washed in cold PBS and fixed in 2% paraformaldehyde in PBS. 121 Fluorescence was determined at the Iowa State University Cell and Hybridoma Facility using a Coulter XL-MCL flow cytometer equipped with standard software. Between 5 x 10^ and 10 x 10^ cells were analyzed for each sample. Live cells were discriminated on the basis of their forward and side scatter parameters. Appropriate compensation was used to optimize dual color analysis. Statistical analyses All experiments were performed at least four times over a one year period. Treatment group means were compared using a repeated measures analysis of variance. Treatment effects were tested with animal(treatment) as the error tenn. Analyses were performed on lymphcyte subset percentages, the percentage of each cell type expressing MHC II or CD25, as well as absolute cell numbers (converted back to either total cells jil""' blood or total cells for each treatment in vitro). Results Confirmation of BLV infection As previously reported, BLV was isolated from the PBMC of all five BLV- inoculated cattle by eight weeks p.i. (Alexandersen et al., 1993). All five BLV- inoculated animals produced antibodies against BLV as detemriined by an agar gel immunodiffusion assay at the beginning of the study (data not shown). None of the 122 control animals tested positive for BLV antibodies prior to or during the study, nor did any of the eight cattle in the study test positive for antibodies to bovine immunodeficiency virus or bovine syncytial virus. Comparison of leukocyte populations in freshly isolated PBMC from BLV-infected vs. control animals As shown in Table 1, total circulating mononuclear cell counts tended to be higher in the BLV-infected group (p = 0.08). In earlier studies of these animals during the first five years p.i., BLV-infected groups had significant increases in numbers of mononuclear cells, B cells, and CD2+, CD4+, CD8+, and WC1+ T cells (Flaming et al., in press, Isaacson et al., submitted). However, with the exception of total mononuclear cell counts, these changes were not consistently present in each of the first five years p.i. The most likely explanation for the lack of statistically significant differences in mononuclear cell numbers in the present study is that only one group of five BLV-infected cattle was examined, therefore the statistical analysis was less powerful. However, a general trend towards increased numbers of both T and B cells continued to be observed. None of the cattle in the BLV- infected group had lymphocyte counts high enough to be classified as persistently lymphocytotic during the study period (Thurmond et al., 1990), although one animal was considered borderiine. There were no significant differences between the groups in either CD4: CDS or T cell: B cell ratios (Table 1). Table 1. Effectsaof BLV Infection on numbers^ of white blood cells and lymphocyte subsets in peripheral blood Group WBC Granulocytes^ Mononuclears^ CD2+ CD4+ CD8+ WC1+ lgM+ CD4:CD8 CD2:lgM Control 5920 2810 3100 1770 1120 388 550 589 2.82 3.50 BLV 8240 3220 5020* 2620 1860 675 501 763 2.84 3.54 a*=p<0.1.'• = p<0.05, "• = p Expression of activation markers CD25 and MHC II on lymphocyte subsets in freshly isolated PBMC Lymphocytes expressing surface CD4, CDS, WC1, or IgM in freshly isolated PBMC from BLV-infected and control cattle were analyzed for expression of immune activation markers MHC II and CD25. As shown In Figure 1 (a), there was a significant increase in the percentage of B cells expressing MHC II in PBMC from BLV-infected cattle (Fig. 1 (a)). Since BLV-infected cattle had increased numbers of circulating B cells, absolute numbers of MHC ll-expressing B cells were also significantly (p < 0.05) increased compared to controls (data not shown). There were no significant differences between the groups in the percentage of T cells expressing MHC II in freshly isolated PBMC (Fig. 1(a)), nor were there any differences between the groups in CD25 expression by any T cell subset (Fig. 1 (b)). A decreased percentage of B cells from BLV-infected animals expressed the CD25 marker. However, because BLV-infected cattle had increased B cell numbers, absolute number of circulating B cells expressing CD25 did not differ significantly between the groups (data not shown). Therefore, circulating B cells from the BLV-infected cattle appeared to be activated with respect to MHC II, but not to CD25. Expression of MHC II by lymphocytes in cultured PBMC PBMC from BLV-infected and control cattle were cultured In the presence or absence of mitogens for either 20 or 68 hours, after which the expression of 125 A. 100 o cO) CO CO £ Q. Control X • (D • BLV OCD (D ££ Q.o CD4 CDS WC1 Lymphocyte phenotype 100 LOoa O o co> CO CO £ & • Control X B BLV o CD a>££ Q. WC1 Lymphocyte phenotype Figure 1. Expression of MHC II and CD25 on lymphocyte subsets in freshly isolated PBMC from BLV-infected cattle (n = 5) and control cattle (n = 3). Data represent means of at least 3 separate sampling dates, analyzed by repeated measures analysis. P values for differences between BLV-infected and control groups are mai1 2(a), in PBMC cultured for 20 hours without exogenous mitogenic stimulation, the percentage of B cells expressing MHC II molecules was increased in the BLV infected group compared to controls, similar to the results for freshly isolated cells. Treatment of PBMC with Con A for 20 hours resulted in MHC II expression by nearly100% of B cells from the BLV-infected group, which was slightly, but significantly, higher than that of B cells from control animals. Con A stimulation also resulted in an increased percentage of CD4+ T cells from BLV-infected cattle expressing MHC II (p < 0.05), as well as a tendency for increased MHC II expression by CD8+ and WC1+ cells (Fig. 2(a)). In contrast, 20 hour treatment with PWM did not increase the percentage of T cells expressing MHC II in the BLV group compared to unstimulated cells. After 68 hours in culture. Increased percentages of CD4+, CD8+, and lgM+ lymphocytes from BLV-infected cattle expressed MHC II molecules, irrespective of the presence of mitogens. Expression of CD25 by lymphocytes in cultured PBMC Figure 3 shows the effect of culturing PBMC on the expression of the Inducible alpha chain of the lL-2 receptor, CD25. There was an increased percentage of B cells expressing CD25 in PBMC from BLV-infected cattle cultured for 20 hours, either with no additional stimulation or in the presence of Con A. 127 A. Unstim ConA PWM 100- CO) CO CO £ • Control & • BLV (DX o c OCD CD Q- ^ CO •»— S S Tt-OOT-2 Q Q Q -& QQOra QQOro O O ^ — oo^— oo§— Lymphocyte phenotype Unstim ConA FWM 100 O X 05 CO CO £ • Control X CD B BLV O CO 2 Q Q O ^ oos—QQOct oo^—OQO^ O O s — Lymphocyte phenotype Figure 2. Effect of BLV infection on MHC II expression by cultured lymphocytes. Data are expressed as percentages of cells in each lymphocyte subset that also expressed MHC II. Data represent means of at least four separate sampling dates, analyzed by repeated measures analysis. PBMC were cultured for either 20 hours (A.) or 68 hours (8). P values for differences between BLV-infected and control groups are marked as follows: * = p<0.1;** = p< 0.05; "* = p < 0.01. 128 However, there were no differences in CD25 expression by any subset of lymphocytes in PBMC cultures treated with PWM. Interestingly, treatment with PWM actually appeared to depress the number of WC1+ T cells of both groups that expressed CD25 in 20 hour cultures compared to unstimulated cell, although no such effect was seen after 68 hours. Discussion The findings reported here add to an accumulating body of data indicating that BLV infection can have immunomodulatory effects in cattle. It had been previously shown that an increased percentage of B cells from BLV-lnfected cattle with PL express the alpha chain of the IL-2 receptor (CD25) following stimulation with mitogens or recombinant IL-2, and an increased percentage of circulating B cells from these cattle spontaneously expressed MHC II (Stone et al. 1994,1995). From these observations, it was suggested that a BLV-induced upregulation of molecules Important in both contact-dependent and cytokine-mediated interactions between B cells and T cells could contribute to the B cell proliferation that is characteristic of PL. In the present study, however, a similar increase in the percentage of B cells expressing MHC II was found in freshly isolated PBMC from BLV-infected animals lacking PL, and an increased percentage of B cells from these BLV-infected, non-PL animals expressed CD25 after culturing for 20 hours, either in the presence or absence of Con A. Therefore, while an upregulated expression of surface molecules Important in B cell-T cell interactions may be necessary for the manifestation of PL, it does not appear to be sufficient. Other 129 Unstim ConA PWM 100- LO oCM o cO) CO CO Control QQOo> QQOto OOOO) oo^— oo^— oo^— Lymphocyte phenotype Unstim ConA PWM 100- LO Qcvj O ca> "eo CO V • Control X B BLV CD U ^ OO •!— S • Lymphocyte phenotype Figure 3. Effect of BLV infection on CD25 expression by cultured lymphocytes. Data are presented as described for Figure 2. 130 factors may also be important In determining susceptibility or resistance to PL. Resistance to PL has been shown to be highly associated with certain MHC (BoLA) types in cattle (Lewin and Bemoco, 1986, Lewin et al., 1988). Although the molecular basis for this resistance is not known, the trait has recently been associated with the presence of the amino acids Glu-Arg at the putative antigen binding region of the BoLA-DFRB3 MHC class II molecule, suggesting a potential role for cell-mediated immune responses in conferring resistance to PL (Lewin, 1994). In addition, a non-immunoglobulin, non-interferon factor capable of blocking BLV transcription is present in the plasma and lymphatic fluid (Gupta et al 182, Gupta and Ferrer 1984, Zandomeni et al., 1992); it is conceivable that lower levels of this factor are present in infected cattle that develop PL compared to those that remain hematologically normal. Thus, cattle that develop PL may be those that are unable to control the proliferation of BLV-infected cells. As described in an accompanying paper, we previously observed that these BLV-infected cattle had enhanced antibody responses to recall antigens compared to uninfected controls. It is likely that this was at least partly due to the overall increase in circulating B and T cell numbers in the BLV-infected group. However, the observation that BLV infection results in an increased percentage of B cells expressing MHC II molecules suggests a possible functional explanation for the enhancement of secondary antibody response of these animals. During the primary response to T cell-dependent antigens, several types of accessory cells have been shown to play a role in presenting antigenic peptides to CD4-t- helper T cells via MHC II molecules. However, B cells have been shown to be very efficient antigen presenting cells during secondary immune responses. The Ig antigen 131 receptor on the B cell surface can capture its specific antigen at very low concentrations, which is then endocytosed, processed, and presented on MHC II molecules to CD4+ T cells. Therefore, in BLV-infected animals, an increase in MHC ll-mediated antigen presentation by B cells would presumably result in increased stimulation of memory CD4+ cells during repeated antigen exposure, which would in turn provide more help for B cell antibody production. Such a mechanism could contribute to the enhancement of secondary antibody responses in BLV-infected animals. In addition to their important function in antigen presentation. MHC II molecules on the surface of B cells are capable of generating intracellular signals that lead to the proliferation of primed B cells (Cambier and Lehmann, 1989). Treatment of murine B cells with F(ab')2 fragments of anti-MHC II monoclonal antibody induced a rise in intracellular cyclic AMP (cAMP) and translocation of protein kinase C (PKC) to the nucleus (Lane et al., 1988). Interestingly, there is evidence that both cAMP and PKC are involved in regulating BLV transcription. The BLV LTR contains several sites that bind to cAMP-responsive elements (Willems et al., 1992), and activation of PKC in BLV-infected cells increases BLV transcription (Jensen et al., 1992). Therefore, the engagement of MHC II molecules could upregulate BLV transcription by increasing the activity of these intracellular signalling molecules. An intriguing possibility is that BLV may have developed a strategy to upregulate MHC II expression in order to enhance its own replication. In humans infected with the human T cell leukemia virus (HTLV), which is closely related to BLV, the HTLV Tax protein has been shown to interact with several host transcription factors, resulting in upregulated expression of both IL-2 and its 132 receptor (Greene et al., 1986), BLV Tax mRNA has been detected in freshly Isolated PBMC from animals with PL or tumors, and to a lesser extent In BLV+, non- PL animals (Haas et al., 1992). In addition, Alexandersen et. al. (1993) reported that the expression level of two products of altemative BLV mRNA splicing, GIV and Rill, in BLV-infected cells was correlated with increased numbers of circulating lymphocytes. It Is unknown at present, however, whether any of these viral products directly Influence MHC II expression, or whether the observed increase in MHC 11 expression may simply result from chronic antigenic stimulation due to the production of BLV antigens. In addition to their presence on classical antigen presenting cells, MHO II molecules are also expressed on T cells in some species. Although the function of MHC II molecules on these cells Is unknown, there is evidence that they are capable of delivering costimulatory signals in human T cells (Odum et al, 1993, Spertini et al., 1992). MHC II expression was upregulated on T cells from BLV- infected, PL cattle following stimulation with mitogens or recombinant IL-2 (Stone, et al., 1995). In the present study, an increased percentage of T cells from BLV- infected, non-PL cattle expressed MHC II molecules after 20 hours in the presence of Con A, or after 68 hours in vitro in the absence of exogenous mitogens. Since BLV provlral DNA has been detected in T cells. It Is possible that the expression of MHC II molecules is a direct effect of BLV Infection, or it may occur subsequent to the activation of BLV-infected B cells In vitro. The increased expression of activation markers on cultured lymphocytes from BLV-infected cattle (CD25 by B cells, MHC 11 by T cells) in the absence of additional mitogenic stimulation is at variance with previous studies showing that mitogenic stimulation is necessary to 133 observe these effects. Compared to previous studies, the longer culture times used in the present study may have allowed the detection of increased T ceil MHC II expression in cultures receiving no mitogens, since the maximum culture time in a previous study was 48 hours (Stone et al, 1995). Altematively, the effects of BLV infection on the expression of activation markers may be qualitatively different in cattle with PL vs. non-PL. Many of the previous studies examining the effects of BLV infection on the host immune system have focused on animals with tumors or PL. In contrast, the experiments reported here and in an accompanying paper showed evidence of enhanced lymphocyte activation in BLV-infected, non-lymphocytotic cattle. As noted earlier, the majority of BLV-infected cattle remain clinically asymptomatic and do not develop PL, at least in the lifespan allotted to most cattle. Although enhancement of humoral immune responses might at first be considered as a potential advantage, it is also possible that an overall increase in B cell activation could reflect a detrimental alteration in normal host immunoregulation. Such a mechanism might contribute to the production losses and increased culling rates that have been associated with BLV infection (Brenner et al., 1989, Wu et al., 1989). Regardless of the practical implications, a better understanding of mechanisms of immune enhancement in BLV-infected cattle will undoubtedly increase our overall understanding of BLV pathogenesis. 134 Acknowledgements The authors wish to thank Tom Skadow for excellent technical assistance. This work was supported by PHS grants R01 CA59125 and AI07378. References Aiexandersen, S., Carpenter, S., Christensen, J., Storgaard, T., Viuff, B., Wannemuehler, Y., Belousov, J. and Roth, J.A., 1993. Identification of altematively spliced mRNAs encoding potential new regulatory proteins in cattle infected with bovine leukemia virus. J. Virol. 67: 39-52. Brenner, J., Van-Haam, M., Savir, D., and Trainin, Z., 1989. The implication of BLV infection in the productivity, reproductive capacity and survival rate of a dairy cow. Vet. Immunol. Immunopathol. 22:299-305. Gambler, J.C. and Lehmann, K.R., 1989. la-mdiated signal transduction leads to proliferation of primed B lymphcytes. J. Exp. Med. 170:877-886. Flaming, K., Van Der Maaten, M., Whetstone, C., Carpenter, S., Frank, D. and Roth, J., 1993. The effect of bovine immunodeficiency-like vims Infection on Immune function in experimentally Infected cattle. Vet. Immunol. Immunopathol. 36:91-105. Flaming, K.P., Frank, D.E., Carpenter, S., and Roth, J.A. Longitudinal studies of immune function in cattle experimentally infected with bovine immunodeficiency- like virus and/or bovine leukemia virus. Vet. Immunol. Immunopathol. In press. Gatel, M.H., Brandon, R.B., Naif, H.M., McLennan, M.W., Daniel, R.C.W. and Lavin, M.F., 1989. Changes In B cell and T cell subsets in bovine leukaemia virus-infected cattle. Vet. Immunol. Immunopathol. 23:139-147. Greene, W.C., Leonard, W.J., Depper, J.M., Nelson, D.L., and Waldmann, T.A., 1986. NIH Conference: The human interieukin-2 receptor: Normal and abnonnal expression in T cells and In leukemias Induced by the human T-lymphotropIc retroviruses. Annals Int. Med. 105:560-72. Gupta, P. and Ferrer, J.F., 1982. Expression of bovine leukemia virus genome is blocked by a nonimmunoglobulln protein in plasma from infected cattle. Science 215:405-407. 135 Gupta, P., Kashmiri, S.V.S. and Ferrer, J.F., 1984. Transcriptional control of the bovine leukemia virus genome: Role and characterization of a non- immunoglobulin plasma protein from bovine leukemia virus-infected cattle. J. Virol. 50:267-270. Haas, L,, Divers, T. and Casey, J.W. 1992. Bovine leukemia virus gene expression in vivo. J. Virol. 66:6223-6225. Jacobs, R.M., Vaili, V.E.O, and Wilkie, B.N., 1981. Response of cows with lymphoma to the intradermal injection of tumor cell antigens and phytohemagglutinln. Can. J. Comp. Med. 45:43-50. Jensen, W.A., Rovnak, J. and Cockerell, G.L., 1991. In vivo transcription of the bovine leukemia virus tax/rex region in normal and neoplastic lymphocytes of cattle and sheep. J. Virol. 65:2484-2490. Jensen, W.A., Wicks-Beard, B.J. and Cockerell, G.L, 1992. Inhibition of protein kinase C results in decreased expression of bovine leukemia virus. J. Virol. 66:4427-4433. Kettmann, R., Cleuter, Y., Mammerickx, M,, Meunier-Rotival, M., Bemardi, G., Bumy, A. and Chantrenne, H., 1980. Genomic Integration of bovine leukemia virus provirus: comparison of persistent lymphocytosis with lymph node tumor form of enzootic bovine leukosis. Proc. Natl. Acad. Sci. USA 77:2577-2581. Lagarias, D.M. and Radke, K., 1989. Transcriptional activation of bovine leukemia virus in blood cells from experimentally infected asymptomatic sheep with latent infections. J. Virol. 63:2099-2107. Lettesson, J.J., Mager, A., Mammerickx, M., Bumy, A. and Delpinchen, A., 1990b. B cells from bovine leukemia virus (BLV) infected sheep with hematological disorders express the CD5 T cell marker. Leukemia 4:377-379. Lane, P.J.L., McConnell, F.M., Schieven, G.L., Clark, E.A. and Ledbetter, J.A., 1988. The role of class II molecules in human B cell activation. Association with phosphatidyl inositol turnover, protein tyrosine phosphorylation, and proliferation. J. Immunol. 144:3684-3692. Lewin, H.A. and Bemoco, D., 1986. Evidence for BoLA-linked resistance and susceptibility to subclinical progression of bovine leukaemia virus infection. Animal Genetics 17:197-207. Lewin, H.A., Wu, M.C., Stewart, J.A and Nolan, T.J., 1988. Association between BoLA and subclinical bovine leukemia virus infection in a herd of Holstein-Friesian cows. Immunogenetics 27:338-344. Lewin, H.A., 1994. Host genetic mechanism of resistance and susceptibility to a bovine retroviral infection. Animal Biotechnology 5:183-191. 136 Mathelse, J.P., Delcommenne, M., Mager, A., Didembourg, C.H. and Lettesson, J J., 1992. CD5+ B cells from bovine leukemia virus infected cows are activated cycling cells responsive to interieukin 2. Leukemia 6:304-309. Odum, N., Kanner, S.B., Ledbetter, J.A. and Svejgaard, A., 1993. MHC Class II molecules deliver costimulatory signals in human T cells throu a functional linkage with IL-2 receptors. J. Immunol. 150:5289-5298. Paul, P.S., Pomeroy, K.A., Johnson, D.W., Muscoplat, C.C., Handwerger, B.S., Soper, F.F. and Sorensen, D.K., 1977. Evidence for the replication of bovine leukemia virus in the B lymphoc^es. Am. J. Vet. Res. 38:873-876. Roth, J.A., Kaeberle, M.L, and Hsu, W.H., 1982. Effects of ACTH administration on bovine polymorphonuclear leukocyte function and lympocyte blastogenesis. Am. J. Vet. Res. 43: 412-416. Schwartz, I., Bensaid, A., Polack, B., Pemn, B., Berthelemy, M. and Levy, D, 1994. In vivo leukocyte tropism of bovine leukemia vims in sheep and cattle. J. Virol. 68:4589-4596. Sordillo, L.M., Hicks, C.R., and Pighetti, G.M., 1994. Altered interleukln-2 production by lymphcyte populations from bovine leukemia virus-infected cattle. Proc. Soc. Exp. Biol. Med. 207:267-273. Spertini, F., Chatila, R. and Geha, R.S., 1992. Signals delivered via MHC Class II molecules synergize with signals delivered via TCR/CD3 to cause proliferation and cytokine gene expression in T cells. J. Immunol. 149:65-75. Stone, D.M., McElwain, T.F. and Davis, W.C., 1994. Enhanced B-lymphocyte expression of IL-2Ra associated with T lymphocytosis in BLV-infected persistently lymphocytotic cows. Leukemia 8:1057-1061. Stone, D., Hof, A.J. and Davis, W.C., 1995. Up-regulation of IL-2 receptor alpha and MHC class II expression on lymphocyte subpopulations from bovine leukemia virus infected cows. Vet. Immunol. Immunopathol. 48:65-76. Taylor, B.C., Stott, J.L., Thumnond, M.A. and Picanso, J.P., 1992. Alteration in lymphocyte subpopulations in bovine leukemia virus-infected cattle. Vet Imm. Immunopathol. 31:35-47. Thumnond, M.C., Carter, R.L., Picanso, J.P. and Stralka, K., 1990. Upper-normal prediction limits of lymphocyte counts for cattle not infected with bovine leukemia virus. Am. J. Vet. Res. 51:466-470. Trainin, Z., Ungar-Waron, H., Meirom, R., Bamea, A. and Sela, M., 1976. IgG and IgM antibodies in nonnal and leukaemic cattle. J. Comp. Path. 86:571-580. 137 Willems, L, Kettmann, R., Chen, G., Portetelle, D., Bumy, A. and Derse, D., 1992. A cyclic AMP-responsive DNA-binding protein (CREB2) Is a cellular transactivator of the bovine leukemia virus long terminal repeat. J. Virol. 66:766-772. Williams, D.L, Amborski, G.F. and Davis, W.C., 1988a. Enumeration of T and B lymphocytes in bovine leukemia virus-infected cattle, using monclonal antibodies. Am J. Vet. Res. 7:1098-1103. Williams, D.L^ Barta, O. and Amborski, G.F., 1988b. Molecular studies of T lymphocytes from cattle infected with bovine leukemia virus. Vet Immunol. Immunopathol. 19:307-323. Wu, M.C., Shanks, R.D. and Lewin, H.A., 1989. Milk cind fat production in dairy cattle Influenced by advanced subclinical bovine leukemia virus infection. Proc. Natl. Acad. Sci. 86:993-996. Zandomeni, R.O., M. Carrera-Zandomeni, M., E. Esteban, E. and J. F. Ferrer, J.F., 1992. Induction and inhibition of bovine leukemia virus expression in naturally infected cells. J. Gen. Virol. 73:1915-1924. 138 CHAPTER 6. SUMMARY AND GENERAL CONCLUSIONS Summary of Experimental Results In this dissertation, experiments have been described that address several aspects of the interactions between bovine immunodeficiency virus (BIV) and/or bovine leukemia virus (BLV) and the bovine immune system. In some lentiviral infections of other species, variations in the kinetics of antibody responses to specific viral proteins are associated with the development of disease. For example, in HIV-infected humans, a loss of antibody reactivity to p24, the major core protein, usually occurs concomitantly with the alterations in lymphocyte functions and decline in CD4 cell counts that acconnpany clinical AIDS (Biggar et al., 1985, Lange et al., 1986, Goudsmit et al., 1987, deWolf et al., 1988). Our hypothesis for the first study was that in cattle experimentally infected with BIV, monitoring of serum antibody to the major BIV core protein may serve as a predictor of the onset of decreased lymphocyte function and the onset of AIDS-like symptoms in cattle. Antibody responses to BIV proteins were studied in experimentally infected cattle (n = 8) by Western immunoblotting using two recombinant BIV fusion proteins; one (Gag3) containing p26, the major core protein of BIV, and another (Env8) containing the amino terminal portion of transmembrane BIV gp42. Although infected cattle developed antibodies reactive with both proteins, antibody reactivity to Gag3 declined to undetectable levels by about six months post-inoculation in seven out of eight cattle. However, none of these animals developed a decline in CD44- T cell counts or evidence of AIDS-like 139 symptoms for at least the next four years. Therefore, unlike human HIV infections, monitoring the levels of antibodies to the BIV major core protein was not a predictor of the onset of clinical disease. In contrast to the declining levels of antibody reactivity to Gag3, antibodies to the Env8 protein persisted in all cattle throughout the four year study period, suggesting that the loss of Gag-specific antibodies was not due to a complete abrogation of viral replication in vivo. Because of this observation, it was next hypothesized that the loss of Gag-specific antibodies was due to an increasing production of BIV p26 antigen in vivo, resulting in the fomnation of p26-containing immune complexes. However, the loss of BIV Gag-specific antibodies was not associated with smy obvious increase in the ease of virus recovery from PBMC. In addition, there were no clinical signs of developing immune complex disease in animals that had lost detectable Gag-specific antibodies. Immunoprecipitation experiments showed that although antibody reactivity to non-denatured GagS declined in a similar manner, there were no detectable levels of circulating p26- containing immune complexes in cattle that had lost Gag-specific antibodies. Therefore, the loss of antibodies did not appear to be attributable to increasing viral replication with subsequent formation of p26-containing immune complexes. Interestingly, BIV could no longer be rescued from the PBMC of one animal (#344) by six weeks p.i., and this was the only animal that maintained detectable levels of Gag-specific antibodies throughout the study. However, since BIV was recovered from the PBMC most of the cattle while they were still Gag-seropositive, the presence of Gag-specific antibodies alone did not appear to prevent virus recovery. 140 Because there was no association between the loss of Gag-specific antibodies and clinically apparent signs of immunosuppression in BlV-infected cattle, the hypothesis for the second study was that a suppression of antigen- specific immune responses played a role in the loss of antibodies in BIV infection. Previous studies of asymptomatic HIV-infected humans revealed a sequential loss of immune reactivity to recall antigens, then to allogeneic MHC, and finally to mitogens (Shearer et. a!., 1986). It was therefore expected that the earliest indicator(s) of BlV-induced immunosuppression would be a decreased responsiveness to a soluble protein antigen and/or to allogeneic MHC, rather than any loss of lymphocyte responses to mitogens, changes in circulating mononuclear cell numbers, or development of overt disease. These parameters wei-e examined in cattle infected for over three years with either BIV only (n = 3), BLV only (n = 5), both BIV and BLV (n = 5), as well as in sham-inoculated controls (n = 3). Because possible interactions had been described for the related human viruses HIV and HTLV, a second hypothesis that was tested in this study was that co-infection with BIV and BLV would alter the effects on immune responses induced by one or both viruses. As obsen/ed in earlier studies of these animals (Flaming et al., 1993, Flaming et al., in press), infection with BIV alone had no consistent effects on numbers of circulating CD4+ T cells, CD8+ T cells, WC1+ gamma-delta T cells, B cells, or monocytes. Lymphocytes from BlV-infected cattle did have significantly reduced proliferative responses to Con A during the fifth year p.i., which supports data obtained from these animals the first four years p.i. showing a tendency for cattle infected with BIV alone to have decreased lymphocyte responses to mitogens, although these earlier differences were not statistically significant 141 (Flaming et al., in press). BIV infection appeared to have little effect on allo- reactivity or on antibody responses to vaccination with tetanus toxoid (TT). Therefore, for this group of cattle and at this stage of infection, the effects of BIV on host immune responses did not resemble those seen in early HIV disease, interestingly, cattle infected with either BLV only, or with BLV and BIV, had increased numbers of circulating B cells and T cells, slightly enhanced lymphocyte proliferative responses to Con A, and significantly enhanced secondary antibody responses to TT, as well as to a second antigen, bovine herpes virus-1. To our knowledge, this elevation of secondary antibody responses to specific antigens in BLV-infected animals has not been previously reported. Since the BLV-infected cattle in these experiments also had increased numbers of circulating lymphocytes, it was unclear whether the obsen/ed enhancement of antibody responses was entirely attributable to increased numbers of antigen-specific Immune cells. Our hypothesis was that in addition to these alterations in cell numbers, BLV infection also causes a lymphocyte activation which may have contributed to the enhancement of acquired immune responses. A third set of experiments was designed to test this hypothesis by comparing the expression of activation markers MHCII and CD25 on freshly isolated and cultured lymphocytes from BLV-infected cattle (n=5) with that of age- matched controls (n=3). Freshly isolated PBMC from BLV-infected cattle were found to have a significantly increased percentage of B cells expressing MHCII molecules compared to non-infected controls. After 20 hours of Con A stimulation in vitro, B cells from BLV-infected animals also had enhanced expression of CD25 (the inducible IL-2 receptor alpha chain) vs. controls Moreover, in PBMC cultured 142 with the T cell mitogen Con A for 20 hours, an increased percentage of T cells from BLV-infected cattle were also found to express MHCII molecules. An increased number of T cells from BLV-infected cattle also expressed MHCII molecules after 68 hours in vitro in the absence of exogenous mitogens. It is therefore likely that BLV-lnduced alterations in lymphocyte function contributed to the enhancement of humoral responses in vivo. Practical Implications The loss of antibody reactivity to BIV Gag proteins in most BlV-infected cattle suggests that assays such as Western blotting, ELISA, or others that may rely on detection of Gag-specific antibodies will tend to underestimate the prevalence of BIV infection. Moreover, since BIV could still be isolated from the PBMC of cattle that had lost detectable Gag-specific antibodies, animals that are seronegative with respect to Gag reactivity would still be expected to be capable of transmitting BIV. Since antibody reactivity to the BIV Env8 recombinant protein persisted In all animals for at least four years post-inoculation, incorporation of this antigen into serological assays should allow a more accurate determination of BLV prevalence rates. Since cattle infected with BIV for over four years did not have any obvious defects in acquired immune responses, experimental BIV infection does not appear to cause the severe degree of immunosuppression seen in Infections with other lentlvimses such as HIV. However, since the incubation period of lentivlral diseases is often much longer than four years, any conclusions about these 143 findings must be qualified by the limited duration of infection. It is also important to note that these cattle were inoculated experimentally with the cell culture-adapted R29 strain of BIV, and lived in a relatively stress- and pathogen-free environment. Although the demonstration of relatively normal acquired immune responses in these animals Is a noteworthy contribution to the body of knowledge about BIV, it provides no information about the importance that such factors as route of inoculation, viral strain, environmental or physiological stress, or infectious cofactors (other than BLV) may play in the pathogenesis of BIV infection. Another important finding of this study was that there were no apparent interactions between BIV and BLV, Since cattle are often found to be infected with both viruses (Whetstone et al., 1990, Snider et al., 1996), any potentiation of the adverse effects of either virus infection would have been a cause for concem. Of course, the same caveats apply to these conclusions as to those in the previous paragraph. One possibility that we did not Investigate, however, was that coinfection with BIV and BLV might facilitate BIV transmission because of the BLV- induced increase in numbers of circulating mononuclear cells. In the course of investigating the effects of coinfection with BIV and BLV, it was discovered that BLV-infected cattle had enhanced humoral responses, particulariy to recall antigens. The most obvious practical implication of this finding is that cattle used in immunological studies of any kind must be screened for BLV infection, since a BLV-induced enhancement of antibody responses would at best contribute to statistical variation, and at worst might lead to erroneous interpretation of experimental results. 144 Theoretical Considerations Because the loss of p24-spectfic antibodies In HIV infection is associated with disease development, the mechanistic basis for this relationship has been investigated. A simultaneous increase in the level of p24 antigenemia is often found in these patients, which had led to the hypothesis that p24-specific antibodies become undetectable because they are bound into circulating immune complexes due to an overwhelming production of antigen (McHugh et al., 1988, Nishanian et al., 1990). However, there is some evidence that the production of p24-specific antibodies is also reduced in some human AIDS patients (Barnes et al., 1988, Portera et al., 1990). In the present study, the loss of antibody reactivity to BIV occun-ed during the first six months post-inoculation. There was no evidence that the antibody loss was attributable to increased viremia, since circulating BIV p26-containing immune complexes were not detectable by immunoprecipitation, nor were there any increases in ability to isolate virus during or subsequent to the antibody loss. Moreover, there was no evidence of immune complex-induced disease in any of these cattle for at least 5 years p.i. Therefore, in contrast to HIV infection, an increasing production of Gag antigens and subsequent formation of Gag-containing circulating immune complexes did not appear to be an important explanation for the loss of Gag-specific antibodies in BlV-infected cattle. It is unknown at present, however, whether the antibody loss in BIV infected cattle is due to a decreased production of Gag-specific antibodies as has been reported in some HIV patients, or perhaps results from a unique mechanism, such as a 145 decreased production of BIV p26 in vivo or a selective loss of p26-specific antibody responses in the continuing presence of p26 antigen production. Despite the prodigious amount of research that has been devoted to BLV over the past 25 years, studies of the Immune responses of BLV-infected cattle to non-BLV antigens have been relatively scarce. Thus, it is perhaps not surprising that the two most important findings of the present studies of the immunological effects of BLV infection have not been previously reported: 1) that BLV-infected cattle can have enhanced humoral responses to recall antigens, and 2) this enhancement may be due in part to a BLV-induced lymphocyte activation, even in cattle that have not developed persistent lymphocytosis. The observation that humoral immune responses were enhanced in BLV+, non-PL cattle is a significant addition to an increasing body of evidence that BLV can cause immune activation in cattle that do not develop tumors. Most previous studies of the effect of BLV on immune functions have focused on cattle with tumors or PL. These studies have shown that an increased number of B cells from BLV+ PL cattle express surface molecules such as CD5, MHCII, and the IL-2 receptor alpha chain, all of which have been implicated as mariners of activated B cells (Possum et al., 1988, Lettesson et al., 1991, Maltheise et al., 1992, Sordillo et al., 1994, Stone et al., 1994, 1995). Other investigators have reported an increased production of IL-2 in supematants of cultured PBMC from BLV+ PL cattle (Sordillo et al., 1994), as well as alterations in amounts of mRNA coding for various cytokines in both PL and non-PL cattle (Pyeon et al., 1996). Thus, the most important implication of the findings presented here is that the BLV-induced immune activation previously observed In vitro may also be manifested as 146 enhancement of humoral responses In vivo. In addition, the fact that enhancement of antibody responses was observed in non-PL cattle implies that it may be a general affect of BLV infection per se, rather than simply resulting from the dramatic increase in circulating B cell numbers seen in cattle with PL. Although the mechanism of BLV-induced enhancement of antibody responses in these animals is unknown, the increased expression of MHCII molecules by B cells in BLV-infected cattle may be a critically important factor. Although B cells are most often thought of as antibody-producing cells, they also serve an important role in processing and presenting antigens to CD4+ T cells, especially during secondary immune responses. The finding that increased numbers of B cells in BLV-infected cattle express MHCII implies that these B cell- CD4+ T cell interactions would be facilitated in vivo. This would be expected to stimulate antigen-specific CD4+ T cells to proliferate and differentiate into cells capable of providing increased help to B cells in temris of soluble cytokines and cell surface costimulatory molecules. This would In tum result in increased differentiation and proliferation of antigen-specific B cells. Therefore, a BLV- induced increase in B cell MHCII expression would be expected to contribute to the increases in lymphocyte numbers and functions that were reported in the present studies. More recent work has demonstrated that imbalances in the amounts of mRNA coding for cytokines such as IL-10, IL-2, and gamma interferon are present in PBMC of BLV-infected cattle with different manifestations of BLV infection (Pyeon et al., 1996). As more bovine cytokine-specific reagents become commercially available, it will be very interesting to detennine whether BLV infection causes host 147 T cells to secrete TH2-type cytokines, which would tend to enhance the proliferation and differentiation of B cells, which are the main targets of BLV. Both BIV and BLV are retrovimses that infect cells of the immune system, such as those found circulating In the peripheral blood. However, the overall findings of the research described in this dissertation suggest that while BIV appears to have a relatively mild Immunosuppressive effect during the early years post-Infection, BLV Infection tends to enhance both numbers and some activities of lymphocytes. Consideration of the different ways In which Antiviruses and oncoviruses utilize their host cell may provide some rationale for the different outcomes of BLV and BIV Infection. Both viruses produce a transactivating protein early In their life cycle that Is Important in regulating viral replication. Although cellular proteins have been shown to influence the replication of both viruses, the transactivating protein of BIV, Tat, can bind directly to the TAR region of BIV RNA independently of the presence of host transcription factors (Chen and Frankel, 1995). Because the BLV Tax protein does not bind directly to the BLV LTR, the presence of cellular factors induced during cell activation is a much more stringent requirement for BLV replication. Therefore, BLV replication may be much more dependent on the activation of Its host cell, while BIV is able to complete Its life cycle with less need to manipulate Its host cell. This may have allowed BIV to evolve Into a relatively non-pathogenic, yet successful virus, while BLV has to work a little harder to succeed. 148 Recommendations for Future Research Mechanism(s) of BIV Gag-specific antibody loss In order to Increase our understanding of host-retrovlnjs Interactions, further research effort should be devoted to elucidating the cause of this unique pattern of antibody loss in BIV infection. Research should focus on other viral and/or host factors that could result in the loss of Gag-specific antibodies, such as whether the expression of BIV Gag proteins is downregulated in vivo at an early stage of viral infection. This could be investigated either by staining of lymphoid tissues for BIV p26 expression at various times post-inoculation, or by developing a quantitiative RT-PCR to amplify BIV Gag-specific RNA from the PBMC of infected animals. In addition. If a sensitive PCR-based assay can be developed for the detection of BIV virions, then the virus load of infected animals could be monitored over time as Gag-specific antibody levels decline . Another possible explanation for the loss of Gag-specific antibodies is that cells that are producing Gag proteins are selectively killed by a BIV Gag-specific cytotoxic T cell (CTL) response. This could provide a selective pressure favoring the sun/ival of BIV strains which express either a mutant Rev protein, or at least much lower levels of Rev which would reduce the amount of unspliced mRNA that is able to code for Gag proteins. The development of assays for the detection of CTL activity against specific BIV proteins would allow the detennination of whether the loss of Gag-specific antibodies correlates with such changes in CTL specficity. 149 A third possible hypothesis is that BIV infection specifically inhibits Gag- specific antibody responses. In order to detemnine whether BIV Gag itself can suppress antibody responses in vitro, cultured PBMC from uninfected cattle could be treated with either BIV Gag3 recombinant protein or with overlapping peptides generated from the reported sequence of BIV p26. Effects of such treatment on lymphocyte proliferative responses to mitogens or antigens, or on expression of MHCII molecules by different mononuclear cell subsets could be determined by flow cytometry. It would be important to use irelevant peptides or recombinant proteins prepared In a similar manner as controls for nonspecfiic effects. In order to further investigate the possibilty that BIV inhibits Gag-specific responses by interfering with antigen presentation, antigen presenting cells (either monocytes or B cells) could be purfied by adherence or panning, treated with Gag3/peptides, then re-mixed with autologous cells and stimulated with antigen to determine the effects on monocyte-lymphocyte or lymphocyte-lymphocyte interactions. Effects of BIV on acquired immune responses The studies described here should be expanded to examine the effects of BIV infection on responses to other adjuvanted and non-adjuvanted antigens, as well as to T cell-independent antigens. In addition, delayed-type hypersensitivity reactions should be measured in BlV-infected cattle following sensitization with either Mycobacterium tuberculosis or M. bovis. In order to more thoroughly characterize the effects of BIV infection under more natural conditions, cattle should be infected with various strains of BIV, including recent Florida isolates (Suarez et 150 al., 1993). Animals should also be inoculated via non-intravenous routes, since natural infection probably can occur through intradermal or mucosal exposure. After varying incubation periods, infected and uninfected cattle should be exposed to environmental and physiological stresses to determine whether the presence of BIV might exacerbate the immunosuppression that is often seen in these situations. Immune responses should also be studied In cattle that have been naturally infected for varying periods, and compared to those of carefully matched controls. Mechanism(s) of BLV-induced enhancement of antibody responses The finding that BLV-infected cattle have enhanced antibody responses to recall antigens should be confirmed in naturally infected cattle, both with and without PL. Naturally infected cattle could be identified in a given herd using the AGIO test, and absolute lymphocyte counts determined for distinguishing cattle with PL. If possible, infected cattle would be separated from uninfected cattle to minimize transmission during the study, and controls would be re-tested at the end of the study to insure that none had seroconverted. Cattle would then be immunized with several different antigens, or different groups could be immunized with different antigens. Both T cell-dependent and T-independent antigens should be used, to better define the role of T cell help in the enhancement of antibody responses. Antibody and lymphocyte proliferative responses could then be compared as In the present study. A recent paper has shown that PBMC from BLV-infected cattle with PL contain higher levels of IL-10 mRNA compared to PBMC from BLV-infected, non-PL 151 cattle or uninfected cattle, while PBMC from the non-PL cattle contained the highest levels of IL-2 mRNA (Pyeon et al., 1996). It may be that BLV-infected cattle tend to develop a TH-2-type pattern of cytokine secretion, which could bias immune responses In favor of antibody production, to the detriment of cell-mediated responses. Consequently, It would be useful to measure either cytokine mRNA, or the production of various cytokines in cultured PBMC, at the same time as antibody responses are being characterized. In addition, delayed-type hypersensitivity responses could be assessed following sensitization of BLV-infected and control cattle to an antigen such as BCG, followed by skin testing. This would allow a determination of whether enhancement of antibody responses is associated with a particular pattem of cytokine production or loss of DTH responses. The potential role of specific BLV proteins, such as Tax, in the enhancement of antibody responses could be investigated in several ways. Cloning and expression of the BLV tax gene would allow the detemnination of whether treatment of uninfected lymphocyte cultures with soluble BLV Tax enhances SLP or expression of CDS, MHCII, IL-2, IL-2R or other molecules important in the control of lymphocyte activation. In addition, cells could be transfected with plasmid constructs where the BLV tax gene is under the expression of an inducible promoter; this would allow the effect of intracellular tax expression to be studied in various cell types. One other interesting possibility is that a BLV protein may act as a superantigen by fomiing a "superantigen bridge" between T cells that express a particular V beta chain and the MHCII molecules expressed on the surface of B cells or other antigen presenting cells. If a BLV protein was identified that causes 152 some proliferation of lymphocytes in vitro, the protein could be added to purified, Irradiated bovine T cells to allow binding to the TCR. These cells could then be washed and added to purified bovine B cells (in separate wells, addition of the BLV protein alone, as well as T cells that had been incubated with an irrelevant protein, would sen/e as controls) to determine whether interactions between the T cell- bound BLV protein and B cell MHCII molecules would result in polyclonal B cell proliferation and Ig production. References Bames, D. M., 1988. Losing AIDS antibodies. Science 240:1407. Biggar, R. J., Melbye, M., Ebbesen, P., Alexander, S., Nielsen, J.O., Sarin, P. and Faber, V., 1985. Variation in human T lymphotropic virus III (HTLV-III) antibodies in homosexual men: decline before onset of illness related to acquired immune deficiency syndrome. Br. Med. J. 291:997-998. Chen, L. and Frankel, A.D., 1994. An RNA-binding peptide from bovine immunodeficiency virus Tat protein recognizes an unusual RNA structure. Biochemistry 33:2708-2715. deWolf, F., Lange, J.M.A, Houweling, J.T.M., Coutinho, R.A., Schellekens, P.T., van der Noordaa, J. and Goudsmit, J., 1988. Numbers of CD4-^ cells and the levels of core antigens of and antibodies to the human immunodeficiency virus as predictors of AIDS among seropositive men. J. Infect. Dis. 158:615-622. Flaming, K., Van Der Maaten, M., Whetstone, C., Carpenter, S., Frank, D. and Roth, J., 1993. The effect of bovine immunodeficiency-like virus infection on immune function in experimentally infected cattle. Vet. Immunol. Immunopathol. 36:91-105. Flaming, K.P., Frank, D.E., Carpenter, 8. and Roth, J.A. Longitudinal studies of immune function in cattle experimentally infected with bovine immunodeficiency like virus and/or bovine leukemia virus. Vet. Immunol. Immunopathol. In press. Goudsmit, J., Lange, J.M.A., Paul, D.A. and Dawson, G.J., 1987. Antigenemia and antibody titers to core and envelope antigens in AIDS, AIDS-related complex, and subclinical human immunodeficiency virus infection. J. Infect. Dis. 155:558-560. 153 Lange, J. M. A., Paul, D.A., Huisman, H.G., de Wolf, F., van den Berg, H., Coutinho, R.A., Danner, S.A., van der Noordaa, J. and Goudsmit, J., 1986. Persistent HIV antigenaemia and decline of HIV core antibodies associated with transition to AIDS. Br. Med. J. 293:1459-1462. McHugh, T. M., Stites, D.P., Busch, M.P., Krowka, J.F., Strieker, R.B. and Hollander, H., 1988. Relation of circulating levels of human immunodeficiency virus (HIV) antigen, antibody to p24, and HIV-containing immune complexes in HIV-infected patients. J. Infect. Dis. 158:1088-1091. Nishanian, P., Huskins, K.R., Stehn, S., Detels, R. and Fahey, J.L, 1990. A simple method for improved assay demonstrates that HIV p24 antigen is present as immune complexes in most sera from HIV-infected Individuals. J. Infect. Dis. 162:21-28. Portera, M., Vitale, F., La Licata, R., Alesi, D.R., Lupo, G., Bonura, F., Romano, N., and Di Cuonzo, G., 1990. Free and antibody-complexed antigen and antibody profile in apparently healthy HIV seropositive individuals and in AIDS patients. J. Med. Virol. 30:30-35. Pyeon, D., O'Reilly, K.L and Splitter, G.A., 1996. Increased interleukin-l 0 mRNA expression in tumor-bearing or persistently lymphocytotic animals infected with bovine leukemia virus. J. Virol. 70:5706-5710. Shearer, G.M., Bemstein, D.C., Tung, K.S.K., Via, C.S., Redfield, R., Salahuddin, S.Z, and Gallo, R.C., 1986. A model for the selective loss of major histocompatibility complex self-restricted T cell immune responses during the development of acquired immune deficiency syndrome (AIDS). J. Immunol. 137: 2514-2521. Suarez, D.L, Van Der Maaten, M.J., Wood, C. and Whetstone, C.A., 1993. Isolation and characterization of new wild-type Isolates of bovine lentivirus. J. Virol. 67:5051- 5055.