Pathobiology of Bovine Leukemia Virus I Schwartz, D Lévy

Pathobiology of bovine leukemia virus I Schwartz, D Lévy

To cite this version:

I Schwartz, D Lévy. Pathobiology of bovine leukemia virus. Veterinary Research, BioMed Central, 1994, 25 (6), pp.521-536. hal-00902257

HAL Id: hal-00902257 https://hal.archives-ouvertes.fr/hal-00902257 Submitted on 1 Jan 1994

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Review article

Pathobiology of bovine leukemia virus

I Schwartz D Lévy

URA-INRA d’Immuno-Pathologie Cellulaire et Moléculaire, École Nationale Vétérinaire dAlfort, 7, avenue du Général-de-Gaulle, 94704 Maisons-Alfort cedex, France

(Received 16 March 1994; accepted 25 July 1994)

Summary ― Bovine leukemia virus (BLV) is a retrovirus similar to the human T-cell leukemia virus (HTLV). Most BLV infected animals (70%) develop a B-cell lymphoproliferative syndrome with altered productive traits and 1 to 5% die with B-cell lymphosarcomas. Although BLV infection is worid-wide, western European countries have almost eradicated it by slaughtering the seropositive animals. BLV infection remains endemic in many countries including the United States and prophylactic strategies involv- ing recombinant vaccine vectors, genetically modified BLV and transgenic animals resistant to the infection are under study.

BLV / animal models / leukosis

Résumé ― Pathobiologie du virus de la leucose bovine enzootique. Le virus de la leucose bovine enzootique (BL V) est un rétrovirus semblable au virus leucémogène humain HTLV La plupart (70%) des bovins infectés par ce virus développent une prolifération lymphocytaire B accompagnée d’une dimi- nution des performances zootechniques, et 1 à 5% des animaux meurent de lymphosarcome B. Bien’ que l’infection par le BL V soit mondialement répandue, les nations de l’Europe occidentale l’ont pra- tiquement éradiquée par abattage systématique des animaux séro-positifs. L’infection par le BLV reste cependant endémique dans de nombreux pays, comme les États-Unis. Des stratégies de pro- phylaxie faisant appel à la vaccination avec des vecteurs vaccine recombinants pour la protéine d’enveloppe, ! des provirus BL V génétiquement modifiés et à des animaux transgéniques résistants sont en cours d’étude.

BLVlmodèles animauxlleucémie

* Correspondence and reprints

INTRODUCTION and molecular knowledge about BLV, and to stress the biological aspects which pro- vide us with an and model The bovine leukemia virus is a cat- original powerful (BLV) for and retroviral diseases. tle retrovirus responsible for bovine enzootic studying fighting leukemia. Its genetic structure, sequence and many similarities pathogenicity present HISTORICAL BACKGROUND to those of the human T-cell leukemia viruses (HTLV-1 and 2). Seventy percent of the infected cattle present a chronic The first reported case of bovine leukemia increase in the number of circulating B lym- was described in 1871 in the Klaipeda area, phocytes, which can lead to a haematolog- Lithuania, and mentioned a cow with super- ical status called persistent lymphocytosis, ficial lymph node hypertrophy and and 1 to 5% of the infected cattle develop a splenomegaly (Burny et al, 1980). There- B-cell lymphosarcoma (Burny et al, 1988). after other cases were reported, indicating’ BLV infection involves a large number of that the disease was propagating west and cattle around the world and the prevalence it eventually reached the USA. In 1917, Ken- is high on the American continent, in Aus- neth showed that the disease was medi- tralia, Africa, and Asia. As herd contamina- ated by a contagious agent (Burny et al, tion follows the introduction of an infected 1980). In 1969, Miller et al discovered that animal, many European countries have lymphocytes of cows with persistent lym- decided to eradicate the infection by apply- phocytosis produced viral particles that were ing strict sanitary measures. Nowadays lym- visible by electron microscopy after in vitro phosarcoma cases are rarely found in most culture (Miller et al, 1969). In 1976, western European countries. Kettmann et al showed that BLV particles Except for the rare lymphosarcoma are RNA exogenous viruses and carry a lesions, the impact of BLV on bovine health RNA reverse transcriptase complex; this and productivity is not well known and still finding allowed the classification of BLV raises many questions. Our purpose here is amongst the oncogenic retroviruses to review the epidemiological, pathological (Kettmann et al, 1976). BLV PATHOGENICITY lymphocyte population of a different lineage from the CD5- B-cells that could be involved in the synthesis of autoantibodies BLV induces a persistent and chronic infec- et This of tion that affects essentially the B lympho- (Hayakawa al, 1985). type lym- is also found in chronic cyte population (Kettmann et al, 1980; Levy phocyte lymphoid leukemia in humans la Hera et al, 1988). et al, 1987). The infection can be transmit- (De Are CD5+ the ted by the cellular components of infected B-lymphocytes preferential for BLV infection? fluorescence bovine blood, by infected cultured cells, or by target Using we BLV free viral particles produced in cell culture activated cell sorting, showed that is found in CD8+ (Burny et al, 1980). Viraemia is only T-lymphocytes, monocytes, detectable during the first 2 weeks of infec- granulocytes, and mostly in B-lymphocytes, tion (Portetelle et al, 1978). Thereafter the with no preference for the CD5+ B-cell pop- The expression of viral antigens is difficult to evi- ulation (Schwartz et al, 1994). param- eters in the of the CD5+ dence but it can sometimes be detected in involved sensitivity to the the intrafollicular and margin zones of lymph B-lymphocytes lymphoproliferative nodes using immunocytochemistry (Heeney potential of BLV remain to be determined. et al, 1992). Cattle develop a serological The ultimate and rare expression of BLV response to capsid and envelope proteins infection is the emergence of a multicentric 2-8 weeks after inoculation (Burny et al, lymphosarcoma that occurs 1-8 years after 1980). These antibodies are lifelong and infection, in about 1-5% of the animals show important level variations that proba- (Burny et al, 1980; Lewin, 1989). Two-thirds bly relate to the physiological and immune of the animals with tumors have had a per- status of the animal. Antibodies to viral reg- sistent lymphocytosis (Burny et al, 1980). ulatory proteins (the Tax and Rex proteins) The tumors may affect 1 or several super- can also be detected in about 20% of ficial and/or deep lymph nodes (Burny et al, infected cattle (Powers et al, 1991). 1980). In a single animal, tumors are mostly In the years following infection, 70% of monoclonal for BLV integration and the virus infected animals show an increase in the does not seem to display preferential inte- B/T lymphocyte ratio in the blood, with a gration sites among tumors from different seemingly normal total number of lympho- animals (Kettmann etal, 1983). According to cytes (4 000 to 10 000/ mm3) (Lewin, 1989). some authors, the tumor cells are pre-B- Among these animals, half develop a per- lymphocytes that are characterized by the sistent lymphocytosis, which corresponds lack of expression of the immunoglobulin to a stable increase in the number of circu- light chain (Van den Broecke et al, 1988); for lating lymphocytes, sometimes reaching up others, they are mature B-lymphocytes to 100 000/mm3 (Lewin, 1989). The inte- showing an isotypic commutation in the gration of the BLV genome is found in 30% immunoglobulin gene (Heeney and Valli, of the circulating lymphocytes and concerns 1990). Most often, the CD5 marker is many clones of lymphocytes (Kettmann et expressed on the tumor cells (Depelchin et al, 1980). The expansion of the lymphocyte al, 1989; Letesson et al, 1991 ). This tumoral population results from the polyclonal pro- form of BLV only affects adult cattle that are liferation of mature B lymphocytes which over 2 years of age, with an incidence peak are cytologically and karyotypically normal around the age of 5-8 years (Lewin, 1989). (Grimaldi ef al, 1983). Most of these lym- This has to be distinguished from the bovine phocytes carry the CDS, CDllb and CD11cc leukemia sporadic cases that affect animals markers (Letesson etal, 1991). In mice, it under 1 year of age and that are not asso- seems that the CD5+ B-cells constitute a B- ciated with BLV infection. BLV TRANSMISSION viral load at the time of infection (Mammer- ickx et al, 1988). It seems that the higher sensitivity of sheep to BLV pathogenicity is Natural transmission related to a higher permissiveness of ovine cells to viral replication (Burny et al, 1980). Ovine intraspecies transmission is extremely In natural conditions, zebus, buf- only cattle, rare in natural conditions and are cases of in falos and have been found to be capybaras utero or accidental transmission et al, infected (Bumy (Burny ef al, 1980). 1980). The virus is transmitted infected by lym- Goats are sensitive to BLV infection, but About 1 500 from phocytes. lymphocytes only 2 out of 24 infected animals developed a cow with persistent lymphocytosis (0.1 !I lymphoid tumors in 10 years (Burny et al, of are sufficient to infect another ani- blood) 1980). mal (Mammerickx et al, 1988). Most of the BLV infection in rabbits is an time, contamination is iatrogenic and occurs interesting model as rabbits an when the animals are manipulated without experimental develop the respect of hygiene recommendations immunodeficiency syndrome 45-763 d after infection et al, After for injections, horning, tattooing or rectal (Altanerova 1989). examinations (Burridge and Thurmond, experimental infection, antibodies to capsid in 3 weeks and the inte- 1981 ). In some areas, transmission by biting protein appear is detectable in insects such as Tabanides has been docu- grated provirus peripheral mented (Oshima et al, 1981 Bronchial blood leukocytes (Altanerova et al, 1989). secretions containing infected lymphocytes In chicken infected at birth, leukemia can be a source of infection especially when development was observed in 4 chickens the animals are maintained in crowded out of 88 and BLV genome could be housing (Burridge and Thurmond, 1981).). detected in the tumoral cells although there BLV does not seem to be sexually trans- was no detectable seroconversion (Altane- missible (Valikhov et al, 1984). Calves can rova et al, 1990). be infected during the second half of ges- Other species, such as rat, pig, rhesus tation, mainly when the mother is in lym- monkey and chimpanzee have been found phocytosis (Burridge and Thurmond, 1981).). to develop antibodies following experimen- Although potentially virulent, the colostrum tal infection, but the integration of BLV is a minor source of infection because the provirus in host cells was not investigated anti-BLV antibodies it contains protect it (Burny et al, 1980). Therefore, one does not (Van der Maaten et al, 1981). know whether the seroconversion represents an immune response to a replicative infection or a serological reaction to the intro- Experimental transmission duction of viral antigens. According to serological studies, humans Sheep have often been preferentially used do not seem to be sensitive to BLV infec- for experimental infection with BLV. Indeed, tion (Oshima et al, 1981 However, human sheep are highly sensitive to BLV infection cells are sensitive to BLV infection in vitro with infected bovine lymphocytes or with (Derse and Martarano, 1990). Besides, viral particles (Mammerickx et al, 1988). hybridization with BLV-derived genomic Sheep develop lymphosarcomas in more sequences was found in 3 cases of human than 90% of the cases in the 6 months to 7 myeloma (Slavikova et al, 1987), and anti- years following infection, depending on the bodies to BLV capsid proteins were found in cases of multiple sclerosis (Clausen et al, of carcasses with tumoral lesions and the 1990). However we should be careful in marking of seropositive animals with culling interpreting these results because different for consumption within 6 months. At that viruses can show crossed genetic and sero- time, farms were also offered a quality label logical reactivities (Maruyama et al, 1989; &dquo;livestock free of enzootic bovine leukemia&dquo;. Battles et al, 1992). More extensive epi- In 1988, measures against BLV were inten- demiological studies should be undertaken sified with a systematic screening for in order to definitively rule out the risk of seropositive animals. In order to encour- infection for consumers and professionals. age breeders to get rid of their seropositive animals, financial grants were given as an allowance of 2 450 French francs per ani- SANITARY AND ECONOMIC IMPACT mal culled within the month following its OF BLV INFECTION identification. Overall these measures cost 394 million francs in 1988. This eradication program was expensive but, in the long run, BLV is not very or infectious in pathogenic it was in order to facilitate the natural conditions. minor economical necessary Only sale of French cattle on the mar- losses are related to BLV-induced tumors. European ket. However, the average dairy production of animals with persistent lymphocytosis is 3-10% lower than that of the herd and it PREVALENCE OF BLV INFECTION can therefore lead to early culling (Brenner IN THE WORLD et al, 1989; Da et al, 1993). Moreover, cattle with lymphocytosis seem less able to get rid of some herd infections (Brenner et al, Commercial exchanges of reproductive cat- 1989), and losses linked to carcass removal tle led to the spread of the infection around when tumors are discovered can be dra- the world. However, the seroepidemiolog- matic at the herd level. In the United States, ical status of the infection remains unknown the economic loss due to BLV infection was in a large number of countries. Moreover, estimated to 86 million dollars in 1992 (Da et regarding the way the virus is transmitted, al, 1993). the infection unevenly affects different Considering the transmission mode of regions within a country, and different farms within a the virus, it is theoretically possible to erad- region. icate the infection using sanitary measures. Denmark (1959) and Germany (1963) In France began a sanitary fight against BLV infection (Burny et al, 1980). In 1981, France had to proceed similarly for the sake of EEC The mean infection rate was 0.075% in commercial exchanges, although seizures 1992, whereas it was 2% in 1981 (Bulletin for lymphosarcoma were only 2.5 per 100 tpid6miologique V6t6rinaire, 2078, June 000. Anyhow, south-western and north- 1993, Minist6re de I’Agriculture et de la eastern France were particularly affected Pbche, Paris, France). However, great vari- regions, with 25.4% of seropositive animals ations remain between regions. In 1992, the (55% of the herds) in the Landes area. The infection rates of the eastern and south- tumoral form of bovine leukemia was western French livestocks were over 5%. declared as a &dquo;legally contagious reputed Therefore, even though the BLV sanitary disease&dquo; in the 8 May 1981 decree. This situation concerning BLV has very much implies the systematic culling and removal improved since 1981, French cattle livestock is not free of enzootic bovine leukemia In Africa yet. In 1992, 10 928 infected cattle were the culled during seroepidemiological Some African countries have estimated the and 2 animals showed screenings only lym- BLV infection rate and came up with a rather phosarcoma lesions (Bulletin Epidemi- high prevalence: 37% in West Africa (Wal- Min- ologique Vétérinaire, 2078, June 1993, rand et al, 1986), 10% in Guinea, 20% in istbre de et de la I’Agriculture Peche, Paris, Zimbabwe, and 50% in some great farms France). in Malawi (Sant6 Animale Mondiale en 1992, Office International des Epizooties, Paris, France). In Europe BLV infection is thus widely spread around the world. Prophylaxis by simple Several European countries established sanitary measures based on culling of the fighting programs against BLV. BLV infection seropositive animals allowed a dramatic has been eradicated in Belgium, Ireland, reduction in the infection rate in western Luxembourg and Norway and is on its way Europe. However, when the infection is to eradication in other western European endemic and affects a majority of animals, countries. It does exist as an enzootic form culling is not applicable and other protec- in eastern European countries such as Alba- tion measures have to be considered. nia, Bulgaria, Yugoslavia and Poland, especially in the Baltic countries (Latvia, Esto- nia, Lithuania) (Sante Animale Mondiale en BLV VIRUS BIOLOGY 1992, Office International des Epizooties, Paris, France). BLV is a type C retrovirus which belongs to the Oncovirinae family. BLV is very similar to human retroviruses HTLV-1 and -2 in terms In North America and Australia of its genetic structure, sequence and regulation of expression. United States, Canada and Australia are strongly affected by BLV infection. In USA and Canada, 66% of the dairy herds (10.2% Viral structure of the animals) and 14% of the meat herds (1.2% of the animals) are infected by the Viral genome virus and there are important variations between states or provinces (Da et al, 1993). The complete sequence of the integrated In North America, bovine leukemia cases BLV is 8 714 nucleotides long (Sagata et do not have to be declared. al, 1985). Like all retroviruses, the extremi- In Queensland, Australia, the prevalence ties of the integrated BLV present a of the infection in the dairy livestock is sequence called LTR (long terminal repeat) 13.7% (Sante Animale Mondiale en 1992, which is composed of 3 consecutive regions Office International des Epizooties, Paris, named U3, R and U5. The U3 region is the France). A sanitary program has been most 5’ region and includes transcriptional established. In New Zealand, 2.4% of the regulatory sequences. The viral mRNAs are herds are infected (Santé Animale Mondiale initiated at the first base of the R region. At en 1992, Office International des Epizooties, least 7 alternately spliced RNAs have been Paris, France). identified as well as 8 open reading frames (orf) designated as gag, ptl, pol, env, tax, Viral structural proteins rex, RIII and GIV (AIeXandersen et al, 1993). The 3’ U3 region includes the polyadenyla- BLV viral particles are constituted of 2 viral tion signals and the 3’ extremity of the R RNAs and of structural proteins that include region corresponds to the polyadenylation the capsid proteins, the envelope proteins, site (fig 1). the reverse transcriptase and the viral protease encoded from the gag, env, pol and Viral mRNAs prt orf respectively. Gag, pol and prt orf products Transcription of BLV genome leads to at least 7 viral mRNAs obtained by alternate The capsid proteins, reverse transcriptase splicing (fig 1). The unspliced mRNA and protease are obtained after cleavage of encodes for the capsid proteins, the reverse 3 different protein precursors initiated from a transcriptase and the viral protease. A common AUG located at the 5’ end of the monospliced viral RNA codes for the enve- gag region. These 3 precursors are made lope glycoproteins, and a doubly spliced from 3 different reading frames, and are the RNA codes for 2 transcription regulatory result of a 1 or 2 nucleotide ribosomal shift proteins called Tax and Rex. The Tax pro- during translation (Haffield et al, 1989). tein stimulates the initiation of transcription, The proteins encoded by the gag gene whereas the Rex protein favors the stabi- are involved in the formation of the viral cap- lization and the processing of the monos- sid (Yoshinaka et al, 1986). A 45 kDa pre- pliced and the unspliced viral mRNAs and cursor is first synthesized and is subse- thus regulates the synthesis of the viral quently cleaved by the viral protease in the structure proteins. Two other mRNAs pre- mature proteins p15, p24 and p12. The p24 sent orf encoding for 2 novel proteins whose is the main capsid protein. The p12 consti- role have yet to be defined, the RIII and GIV tutes the nucleocapsid and is associated proteins. with 2 viral RNAs. The p15 is a myristilated protein and is associated both to the cap- Finally, unlike lentiviruses, envelope pro- sid proteins and to the lipid bilayer of the teins, the BLV gp30 and gp51 are well con- viral membrane (Yoshinaka et al, 1986). served among the different variants studied. The between the env The mature pol product (70 kDa) is pro- comparison nucleotide of 7 Amer- duced from a 145 kDa precursor. It presents sequences European, ican and BLV isolates a a RNA-dependent DNA polymerase activity Japanese showed lower than 6% et (reverse transcriptase) at its N-terminal por- variability (Mamoun al, whereas the nucleotide tion and an RNAase H activity at its C-ter- 1990) variability reaches around 66% between the env minal portion. The pol product also has an pro- integrase activity (Burny etal, 1980). teins of HIV-1 isolates (Goudsmit et al, 1991). The prt is synthesized from a 65 kDa precursor which leads after cleavage to a 14 kDa dimeric It aspartic protease. performs Regulation of BLV expression the cleavage of the gag, prt, pol and env precursors (Katoh et al, 1989a). The 3’ portion of the BLV genome, formerly Env orf products called the pX region, codes for at least 4 the Tax, Rex, Rill and GIV The mature and proteins, pro- gp51 gp30 glycoproteins teins; it is actually well documented that the are obtained from the cleavage of a 72 kDa Tax and Rex proteins participate in the reg- glycosylated precursor the viral by protease. ulation of BLV expression. Gp51 is found at the surface of the viral membrane and ensures the recognition of the cellular viral receptor (Vonbche et al, Activation of BLV expression 1992a). The N-terminal segment of gp30 is by the Tax protein inserted in an oblique orientation into the lipid bilayer and anchors the gp51/gp30 BLV Tax protein is a 34 kDa nuclear phos- complex in the viral envelope and the phoprotein which transactivates the initia- infected cell membrane (Von6che et al, tion of RNA synthesis from the U3 region 1992b). The gp30/gp51 association helps of the viral LTR (Katoh et al, 1989b). The membrane fusion during infection of the host U3 region includes three 21 base pair (bp) cell and syncitia formation (Voneche et al, elements that are responsive to Tax trans- 1992a). activation (Derse, 1987). These 3 elements are and include an 8- The gp30 intracytoplasmic portion pre- strongly homologous which is the cAMP- sents a motif (tyrosine X X leucine), which is bp core, homologous to found in the signal transduction molecules of responsive element (CRE). A mutation in the T cell receptor (Ç chain) and of the B the CRE in the BLV LTR abolishes trans- cell receptor (iga and Ig(3 chains). The stim- activation by the Tax protein (Katoh et al, ulation of a chimeric molecule made of the 1989b). In addition, a cell nuclear factor CD8 extracytoplasmic portion and the gp30 called CRE-binding-protein-2 (CREB2) has intracytoplasmic domain with anti-CD8 anti- been shown to bind to the 21 bp regions bodies induces the phosphorylation of cel- and to be implicated in the transactivation of the BLV LTR lular proteins and a calcium influx into the (Willems et al, 1992b). cell (Alber et al, 1993). Thus, gp30 may The U3 region of HTLV-1 also includes have a signalling function in the infected cell three 21 bp repeats centered on CRE, which and anti-BLV envelope antibodies may be are homologous to that found in the BLV able to stimulate gp30 and dysregulate the LTR; these are essential for HTLV-1 trans- infected cell proliferation. activation by the HTLV-1 Tax protein as well as for promoter basal activity (Jeang et al, for 2 novel proteins, the Rill and GIV pro- 1988). Neither of the HTLV-1 or BLV Tax teins (Alexandersen et al, 1993). The proteins are bound directly to DNA; this expression level of the GIV RNA correlates implies that cellular nuclear factors are with the development of persistent lympho- needed to mediate transactivation (Jeang cytosis. However these 2 proteins have not et al, 1988). In the case of HTLV-1, the Tax been detected in vivo and their role is still protein makes a nuclear complex with CREB under study. Using expression vectors for and with activating transcription factor-2 GIV and RIII in cell transfection experiments, (ATF-2) and increases CREB/ATF-2 affinity it could be shown that GIV weakly transac- for CRE in the 21 bp units but not for the tivates BLV LTR activity and potentiates other CRES, indicating that the CRE flanking Rex protein function whereas RIII inhibits Rex function Work- nucleotides are important in Tax transacti- protein (S Alexandersen, vation activity (Franklin et al, 1993). The shop on the Pathogenesis of Animal Retro- La CRE flanking nucleotides may also govern virus, Rochelle, France, 1993). the specificity of the Tax proteins for their respective LTR as the Tax protein of BLV Lymphocyte activation is unable to transactivate HTLV-1 LTR and and viral synthesis vice versa (Derse, 1987). In vivo mRNAs coding for Tax and/or Rex Regulation of BLV expression proteins are easily detectable by the highly by the Rex protein sensitive technique of reverse transcription followed by polymerase chain reaction (RT- Rex is a 18 kDa nuclear phosphoprotein PCR) (Haas et al, 1992). Unspliced or which is involved in the post-transcriptional monospliced viral mRNAs are absent or control of BLV expression. The Rex protein weakly expressed, a finding that fits with allows the accumulation of unspliced or the paucity of virus expressing cells detected monospliced viral RNAs, probably by enforc- by immunocytochemistry (Jensen et al, ing their stability (Derse, 1988). A 250-bp 1990). BLV infection is therefore rarely sequence located between the polyadeny- replicative in vivo and the mechanisms that lation signal and the polyadenylation site in regulate this viral latency are still elusive. the 3’ R region of the LTR, is necessary for However a non-immunoglobulin plasma fac- Rex function (Derse, 1988). This sequence, tor absent from serum has been shown to called Rex responsive element (RRE), as participate in the inhibition of the viral syn- well as its secondary structure, present great thesis in vivo (Gupta et al, 1984). homologies with the equivalent HTLV-1 The culture of infected lymphocytes from structure. the HTLV-1 Rex Interestingly pro- sheep or cattle with lymphocytosis unblocks tein can the BLV Rex and replace protein the viral synthesis after 6 h incubation at vice versa (Felber et al, In the 1989). 37°C (Cornil et al, 1988). In our laboratory, absence of Rex protein, only spliced doubly we have shown that BLV structural protein RNAs are which that synthesized, implies is stimulated the of virion is condi- synthesis strongly following expression proteins the activation of T lymphocytes by lectins tioned by Rex protein. such as concanavalin A or phytohaemag- glutinin A (Djilali et al, 1987; Cornil et al, Rill and GIV proteins 1988). This stimulating effect is mimicked by the supernatant of activated lymphocyte The study of BLV mRNA differential splicing cultures which indicates that cytokines are reveals the existence of 2 mRNAs coding implicated in the stimulation of viral synthesis (Cornil et al, 1988). The identity of the (Lilienbaum et al, 1990), the transforming cytokine(s), as well as the cell type respon- growth factor p (Kim et al, 1990). The trans- sible for virion synthesis (B cells, CD8+ T- activation of cellular genes is proposed in cells, monocytes or granulocytes) are order to explain the lymphocyte perturba- unknown (Djilali et al, 1987). tions induced by HTLV-1. The BLV Tax protein can transactivate the c-fos and somato- statin gene promoters (Katoh et al, 1989b). BLV and cell transformation Recent studies have also shown that the Tax transactivating domain (a zinc finger The mechanism by which BLV leads to region) is different from the domain involved tumoral development has not yet been elu- in cell transformation (Willems et al, 1992a). in BLV cidated . No rearrangement of any known Therefore, the way which Tax protein to involve a oncogene (c-myc, c-myb, c-erbA, c-erbB, leads oncogenesis may c-src, c-Ha-ras, c-fos and c-sis) could be function other than transactivation. For evidenced in BLV-induced tumors instance, viral antigens, such as the SV40 (Kettmann et al, 1983). Moreover, BLV does T antigen (Fanning and Knippers, 1992), not show any clear p.referential integration the adenoviral E1A antigen (Whyte et al, site between tumors, a finding which does 1989) and human papillomavirus E7 (Dyson not support the insertional mutagenesis et al, 1989), bind to cellular proteins and hypothesis (Kettmann et al, 1983). How- modulate or abrogate their functions. The ever, a number of experimental observa- cellular proteins that physically associate tions indicate that BLV Tax protein, as well with Tax during cell transformation are as HTLV-1 Tax protein, can act as onco- actively sought and may reveal the exis- genes. Transfection of BLV Tax together tence of new molecules involved in cell with the Ha-ras oncogene renders rat transformation. embryo fibroblasts tumorigenic in nude mice Although Tax is most probably involved in (Willems et al, 1990). Moreover, mRNAs the development of BLV-induced lympho- coding for Tax (and/or Rex) are detectable proliferative syndrome, other viral proteins in animals with persistent lymphocytosis or namely Rex, GIV, Rill and even env pro- with tumors whereas mRNAs coding for teins, are also likely to play a role in BLV structural proteins are not (Haas et al, 1992). pathogenicity. The Rex protein of BLV may The transactivating protein Tax is therefore disturb cellular mRNAs stability, as the Rex the most probable viral element involved in of HTLV-1 increases the IL-2 receptor lymphocyte transformation. However, its mRNA half-life (White et al, 1991 The env expression is not always encountered in protein may transduce intracellular signals advanced tumoral stages (Sagata et al, after stimulation by anti-env antibodies 1985); Tax could therefore play a role in the (Alber et al, 1993). The possible role of RIIIl l initiation of tumorigenesis but does not seem and GIV proteins in BLV pathogenicity must to be essential in the maintenance of the be investigated, especially for GIV whose tumoral stage. mRNA level is high during persistent lym- In the case of transformation with HTLV- phocytosis (Alexandersen et al, 1993). 1, a number of cellular genes transactivated Therefore, lymphocyte dysregulation by Tax have been identified, such as the induced by BLV infection potentially impli- interleukin-2 receptor (Inoue et al, 1986), cates several viral molecules and a lot has the granulocyte/macrophage-colony stim- still to be done in order to delineate the spe- ulating factor (Nimer et al, 1989), vimentin cific role of each of them. PROPHYLAXIS OF BLV INFECTION BLVs. BLV mutants with deletions in various AND PATHOGENICITY genes have been constructed (Willems et al, 1993). Since BLV proviruses injected As mentioned above, it is important to intradermally in cationic liposomes are infec- develop fighting strategies against BLV in tious in sheep, the infectious and pathogenic countries where the spread of the infection potentials of mutant proviruses could be tested The infectious is too high to undertake a sanitary prophy- (Willems etal, 1993). laxis. Moreover, BLV represents a good ani- potential is altered for gag, pol and envv mal model for studying the prophylaxis of mutated proviruses but not for mutants in retrovirus diseases. Indeed, its complex RIII and GIV. The pathogenic potential of genetic structure is analogous to the one of RIII and GIV mutants remains to be estab- primate retroviruses, and its pathogenicity in lished. In case these mutants are not sheep is reproducible, frequent and easily pathogenic, they may be used as infectious detectable within delays suitable to experi- but protective vectors against a wild type mentation. Classical vaccination strategies virus. have been tested. Innovative strategies via For the same purpose, a chimeric molecular targeting of the viral regulatory provirus made of BLV gag, pol and env proteins are under study and are discussed genes under the control of heterologous below. LTRs derived from the murine spleen necro- sis virus was constructed and turned out to be replicative in vitro (Temin, 1993). The Vaccination with env recombinant infectiosity and protection conferred by this vaccine chimeric virus devoid of regulatory proteins remain to be tested. The humoral immune response to BLV is protecting since serum immunoglobulins of infected sheep prevent infection of healthy Selection of naturally resistant animals sheep in an intradermal challenge with infected lymphocytes (Kono et al, 1986). A lmmunogenetic studies have shown that env recombinant vaccine vector was con- class-1 major histocompatibility complex structed and tested a team by Belgian alleles are associated with resistance or et al, 1991, The (Portetelle 1993). protec- sensitivity to the development of a persis- tion was tested in with an intravenous sheep tent depending on the race of lymphocytosis, challenge infected lymphocytes 6 weeks of the cattle. For instance, in Holstein herds, after vaccinal boost. The vector protecting a relative sensitivity to persistent lympho- codes for the and On gp51 gp30 proteins. cytosis is associated with the BoLA w8-1 the other hand, immunization with vectors allele (Lewin et al, 1988). On the other for or alone does not coding gp51 gp30 hand, in Shorthorn herds, it was found to induce antibodies and is not neutralizing be associated with the BoLA DA 12-3 allele The immune induced protective. memory (Lewin, 1986). However the development by this vaccine remains to be studied. of persistent lymphocytosis is strongly associated with the DRB2 and DRB3 alleles of the class-2 major com- Vaccination with a BLV histocompatibility non pathogenic plex (Van Eijk et al, 1992; Xu et al, 1993). These reports indicate that major histo- This strategy relies on the construction of compatibility complex genes are associated infectious, immunogenic, but non-pathogenic with sensitivity to BLV, but more extensive studies are needed to find the genes directly proteins, which would quench the wild type responsible. Information on these genes Tax and Rex proteins, thereby blocking viral may have agronomic applications shedding replication. These cattle would then be resis- light on genetic resistance to retroviral infec- tant to the propagation and pathogenicity tions and/or cancer development. of the virus. It is clear that such transgenic animals would only be experimental animals that Construction of resistant transgenic would help to evaluate the efficiency of these animals &dquo;gene prophylaxis&dquo; molecules. Beside the cost of these transgenic constructions, their Today transgenesis in cattle is possible. introduction to the field is not easy. It is nec- Constructing transgenic animals that essary to keep in mind the potential risks inherent to in express molecules aimed at conferring resis- exogenous gene expression because the and zootech- tance to BLV is conceivable. In this per- animals, sanitary nical are difficult to spective, a ribozyme that cleaves the consequences predict. Tax/Rex coding RNA has been constructed Nevertheless, their introduction in some American farms where the infection is (Cantor et al, 1993). Ribozymes are com- could be considered. plementary RNA sequences to a target RNA widespread seriously and are made of a hammer-shaped region endowed with a catalytic activity. An effi- CONCLUSION cient ribozyme to the RNA coding for BLV Tax and Rex proteins was shown to reduce the reverse transcriptase activity by 92% in BLV infection is widespread throughout the a BLV-infected bat lung cell line (Cantor et world and it significantly affects cattle al, 1993). When expressed constitutively in zootechnical performances without any dra- transgenic animals, such a ribozyme would matic health damage. However, we still do prevent the accumulation of Tax and Rex not know whether BLV potentiates other coding RNAs from the very beginning of herd infection symptomatologies, especially infection and would thus interfere with Tax in the USA and in Africa where BLV infection transactivation. Consequently, the virus is endemic. In any case, it would be inter- infection of the neighboring sensitive cells esting to propose a cheap prophylaxis would be considerably slowed down or even related to the BLV-associated loss of abrogated and the transgenic animals may income. For the time being, vaccination then control the infection and resist to the strategies are still under trial and are not virus pathogenicity. yet applicable on a large scale. In a similar perspective, transdominant Furthermore, BLV infection in sheep turns mutants for HTLV-1 Tax and Rex proteins out to be an excellent experimental model have also been obtained (Rimsky et al, for studying the complex primate retro- 1989; Gitlin et al, 1991 These mutants viruses, as no equivalent model can be pro- are inactive in transactivation and they inhibit posed with the murine or avian retroviruses wild type Tax and Rex transactivation, prob- whose genomic structures are much sim- ably by interfering with their transport to the pler. BLV pathogenicity is reproducible and nucleus. Isolation of such BLV Tax and Rex readily detectable in sheep, an animal eas- mutants could be very interesting for con- ily amenable to experimentation. It is also structing transgenic BLV-resistant animals. a good model for analyzing the molecular Transgenic cattle would constitutively events involved in cell transformation and express transdominant mutant Tax and Rex tumor progression. However many questions remain open. Why are only CD5+ B Cantor GH, McElwain TF, Birkebak TA, Palmer GH cleaves rexltax mRNA and inhibits sensitive to the (1993) Ribozyme lymphoproliferative poten- bovine leukemia virus expression. Proc Natl Acad tial of BLV in vivo? What is the real impact Sci USA 90, 10932-10936 of the Tax protein on lymphocyte dysregu- Clausen J, Hoff-Jorgensen R, Rasmusen HB (1990) lation? With which molecules does it inter- Antibody reactivity against animal retrovirus in mul- act? The answers to these many questions tiple sclerosis. Acta Neurol Scand 81, 223-228 will allow us to unveil the cellular molecules Cornil I, Delon P, Parodi AL, Levy D. (1988) T-B-cell cooperation for bovine leukemia virus expression in that are essential for the control of lympho- ovine lymphocytes. Leukemia 2, 313-3177 homeostasis. cyte Da Y, Shanks RD, Stewart J, Lewin HA (1993) Milk and fat yields decline in bovine leukemia virus-infected Holstein cattle with persistent lymphocytosis. Proc ACKNOWLEDGMENT Nati Acad Sci USA 90, 6538-65411 De la Hera A, Alvarez Mon M, Sanchez Madrid F, Mar- tinez C, Durantez A (1988) Co-expression of Mac- The authors are especially grateful to B Schwartz 1 and p150,95 on CD5+ B cells. Structural and func- for critical reading of the manuscript. tional characterization in a human chronic lymphocytic leukemia. Eur J Immunol 18, 1131- 1134 REFERENCES Depelchin A, Letesson JJ, lostrie Trussart N, Mamm- erickx M, Portetelle D, Burny A (1989) Bovine leukemia virus (BLV)-infected B-cells express a Alber G, Kim KM, Weiser P, Reisterer C, Carsetti R, marker similar to the CD5 T cell marker. Immunol Reth M (1993) Molecular mimicry of the antigen Lett 20, 69-76 receptor signalling motif by transmembrane proteins Derse D (1987) Bovine leukemia virus transcription is of the Epstein-Barr virus and the bovine leukemia controlled by a virus encoded trans-acting factor virus. Current Biology 3, 333-338 and by cis-acting response elements. J Virol 61, Alexandersen S, Carpenter S, Christensen J et al (1993) 2462-2471 Identification of altematively spliced mRNAs encod- Derse D (1988) Trans-acting regulation of bovine ing potential new regulatory proteins in cattle infected leukemia virus mRNA processing. J Virol62, 1115- with bovine leukemia virus. J Viro167, 39-52 11199 Altanerova V, Ban J, Altaner C (1989) Induction of Derse D, Martarano L (1990) Construction of a recom- immune deficiency syndrome in rabbits by bovine binant bovine leukemia virus vector for analysis of leukemia virus. AIDS 3, 755-758 virus infectivity. J Virol64, 401-405 Altanerova V, Ban J, Kettmann R, Altaner C (1990) Djilali S, Parodi AL, Levy D (1987) Bovine leukemia of in Induction leukemia chicken by bovine leukemia virus replicates in sheep B-lymphocytes under a T- virus due to insertional mutagenesis. Arch Geschwul- cell released factor. Eur J Cancer Clin Oncol23, 81- sfforsch 60, 89-96 85 Battles JK, Hu MY, Rasmussen L, Tobin GJ, Gonda MA Dyson N, Howley PM, Mungel P, Harlow E (1989) The (1992) Immunological characterization of the gag human papilloma virus 16 E7 oncoprotein is able to of bovine virus. J gene products immunodeficiency bind to the retinoblastoma gene product. Science Virol66, 6868-6877 341,934-937 Brenner J, Van Haam M, Savir D, Trainin Z (1989) The Fanning E, Knippers R (1992) Structure and function implication of BLV infection in the productivity, repro- of simian virus 40 large tumor antigen. Ann Rev ductive capacity and survival rate of a dairy cow. Biochem 61, 55-85 Vet Immunol Immunopathol 22, 299-305 Felber BK, Derse D, Athanassopoulos M, Campbell M, A, Bruck C, Chantrenne H et at Bovine Burny (1980) Pavlakis GN (1989) Cross activation of the rex pro- Leukemia Virus: Molecular and Biology Epidemiol- teins of HTLV-1 and BLV and of the rev protein of Raven New York ogy, Press, HIV-1 and non-reciprocal interactions with their RNA Burny A, Cleuter Y, Kettmann R et al (1988) Bovine responsive elements. New Biologist 1, 318-330 leukemia: facts and hypotheses derived from the Franklin AA, Kubik MF, Uittenbogaard MN et al (1993) study of an infectious cancer. Adv Vet Sci Comp Transactivation by the human T-cell leukemia virus Med 32, 149-170 Tax protein is mediated through enhanced binding of Burridge MJ, Thurmond MC (1981) An overview of activating transcription factor-2 (ATF-2) ATF-2 modes of transmission of bovine leukemia virus. In: response and cAMP element-binding protein Proc 85th Ann Meet US Anim Hlth Ass, 165-169 (CREB). J Biol Chem 268, 21225-21231 Gitlin SD, Lindholm PF, Marriott SJ, Brady JN (1991) Kettmann R, Portetelle D, Mammerickx M et al (1976) Transdominant human T-cell lymphotropic virus type Bovine leukemia virus: an exogenous RNA onco- I TAXI mutant that fails to localize to the nucleus. J genic virus. Proc Natl Acad Sci USA73, 1014-10188 Virol65, 2612-2621 Kettmann R, Cleuter Y, Mammerickx M et al (1980) Goudsmit J, Bac, NKT, Nara PL (1991) Genomic diver- Genomic integration of bovine leukemia virus : com- sity and antigenic variation of HIV-1: links between parison of persistent lymphocytosis with lymph node pathogenesis, epidemiology and vaccine develop- tumor form of enzootic bovine leukosis. Proc Natl ment. FASEB J5, 2427-2436 Acad Sci USA 77, 2577-2581 Grimaldi MG, Poll G, Sartoelli P, Caldora C, Oldani L, Kettmann R, Deschamps J, Claustriaux JJ, Palm, R, Bumy Locatelti A (1983) Karyotype analysis of lympho- A (1983) Chromosome integration domain for bovine cytes from cattle at different stages of bovine leukemia provirus in tumors. J Virol47, 146-150 leukemia virus infection. Br Vet J 139, 240-246 Kim SJ, Kehrl JH, Burton J et at (1990) Transactivation Gupta P, Kashmiri S, Ferrer J (1984) Transcriptional of the transforming growth factor beta 1 (TGF-beta 1) control of the bovine leukemia virus genome: role gene by human T lymphotropic virus type 1 tax: a and characterization of a non-immunoglobulin plasma potential mechanism for the increased production protein from bovine leukemia virus infected cattle. of TGF-beta 1 in adult T cell leukemia. J Exp Med J Virol50, 267-270 172, 121-129 Haas L, Divers T, Casey W (1992) Bovine Leukemia Kono Y, Arai K, Sentsui H, Matsukawa S, Itohara S Virus gene expression in vivo. J Virol66, 6223-6225 (1986) Protection against bovine leukemia virus infection in active and immuniza- Hatfield D, Feng Y, Lee B, Rein A, Levin J. Oroszlan S sheep by passive tion. J Vet (1989) Chromatographic analysis of the aminoacyl Jpn Sci 48, 117-125 tRNAs which are required for the translation of codon Letesson JJ, Van den Broecke A, Marbaix Cleuter Y et at and around the ribosomal frameshift sites of HIV, al (1991) FACS analysis of bovine leukemia virus HTLV-1 and BLV. Virology 173, 736-754 (BLV)-infected cell lines with monoclonal antibodies to B cells and to Hayakawa K, Hardy RR, Herzenberg LA (1985) Pro- (mAbs) monocytes/macrophages. Vet Immunol lmmunopathol 27, 207-2133 genitors for Ly-1 B cells are distinct from progeni- tors for other B cells. J Exp Med 161, 1554-1568 Levy D, Kettmann R, Marchand P, Djilali S, Parodi AL Heeney JL, Valli VEO (1990) Transformed phenotype (1987) Selective tropism of bovine leukemia virus ovine of enzootic bovine reflects differentiation (BLV) for surface immunoglobulin-bearing lym- lymphoma Leukemia linked leukemogenesis. Lab Invest 62, 339-346 phocytes. 1, 463-465 Heeney JL, Valli PJS, Jacobs RM, Valli VEO (1992) Evi- Lewin HA (1986) Evidence for BoLA-linked resistance to clinical of bovine dance for bovine leukemia virus infection of periph- and susceptibility progression leukemia virus infection. Anim Genet 17, 197-207 eral blood monocytes and limited antigen expression in bovine lymphoid tissue. Lab Invest 66, Lewin HA, Wu MC, Stewart JA, Nolan TJ (1988) Asso- 608-6177 ciation between BoLA and subclinical bovine Inoue J, Seik, M, Taniguchi T, Tsuru S, Yoshida M (1986) leukemia virus infection in a herd of Holstein-Friesian cows. 27, 338-344 Induction of interleukin 2 receptor gene expression Immunogenetics by p40x encoded by human T-cell leukemia virus Lewin HA (1989) Disease resistance and immune type 1. EMBO J 5, 2883-2888 response genes in cattle: strategies for their detection and evidence of their existence. J Jeang OT, Boros I, Brady J, Radanovich M, Khoury G Dairy Sci72, 1334-1348 (1988) Characterization of cellular factors that inter- act with the human T-Cell Leukemia Virus Type 1 Lilienbaum A, Duc Dodon M, Alexandre C, Gazzolo L, p40X responsive 21 base-pair sequence. J Virol62, Paulin D (1990) Effect of human T-cell leukemia 4499-4509 virus type I tax protein on activation of the human Jensen WA, Sheehy SE, Fox MH, Davis WC, Cockerell vimentin gene. J Virol64, 256-263 GL (1990) In vitro expression of bovine leukemia Mammerickx M, Palm R, Portetelle D, Burny A (1988) virus in isolated B lymphocytes of cattle and sheep. Experimental transmission of leukosis in sheep: Vet Immunol Immunopathol26, 333-342 latency period of the tumoral disease. Leukemia 2, Katoh I, lkawa Y, Yoshinaka Y (1989a) Retroviruses 103-107 protease characterized as a dimeric aspartic pro- Mamoun RZ, Morisson M, Rebeyrotte N et al (1990) teinase. J Virol63, 2226-2232 Sequence variability of bovine leukemia Virus env and its relevance to the structure and anti- Katoh 1, Yoshinaka Y, lkawa Y (1989b) Bovine leukemia gene of the J 4180-4188 virus trans-activator p38tax activates heterologous genicity glycoproteins. Virol64, promoters with a common sequence known as a Maruyama K, Fukushima T, Mochizuki S (1989) Cross- cAMP-responsive element or the binding site of a reactive antibodies to BLV and HTLV in bovine and cellular transcription factor ATF. EMBO J 8, 497- human hosts with retrovirus infection. Vet Immunol 503 Immunopathoi22, 265-273 Miller JM, Miller LD, Olson C, Gillette KG (1969) Virus- Van der Maaten MJ, Miller JM, Schmerr MJ (1981) Effect like particles in phytohemagglutinin, stimulated lym- of colostral antibodies on Bovine Leukemia Virus phocyte cultures with reference to bovine lym- infection of neonatal calves. Am J Vet Res 42, 1498- phosarcoma. J Ntl Cancer Inst 43, 1297-1305 1500 Nimer SD, Gasson JC, Hu K et al (1989) Activation of the Van Eijk MJT, Stewart JA, Beever J, Femando RL, Lewin GM-CSF promoter by HTLV-I and -II tax proteins. HA (1992) Development of persistent lymphocytosis Oncogene 4, 671-676 in cattle is closely associated with BRB2. Immuno- Oshima K, Okada K, Nomakunai S, Yoneyma Y, Sato B, genetics 37, 64-68 Takahashi I (1981) Evidence of horizontal trans- Voneche V, Callebaut I, Kettmann R, Brasseur R, Bumy mission of bovine leukemia virus due to blood-suck- A, Portetelle D (1992a) The 19-29 amino acid frag- ing tabanid flies. Jpn J Vet Sci 43, 79-81 ment of GP51 adopts an amphiphilic structure and Portetelle D, Bruck C, Bumy A, Dekegel D, Mammer- plays a key role in the fusion events induced by ickx M, Urbain J (1978) Detection of complement bovine leukemia virus. J Biol Chem 267, 15193- dependent lytic antibodies in sera from bovine 15197 leukemia virus infected animals. Vet Res 9, 667-674 Vonbche V, Portetelle D, Kettmann R et al (1992b) Fuso- Portetelle D, Limbach K, Bumy etal(1991) Recombi- genic segments of bovine leukemia virus and simian nant vaccine virus expression of the bovine leukaemia immunodeficiency virus are interchangeable and virus envelope gene and protection of immunized mediate fusion by means of oblique insertion in the sheep against infection. Vaccine 9, 194-200 lipid bilayer of their target cells. Proc Natl Acad Sci USA 89, 3810-3814 Portetelle D, Callebaut I, Bex F, Burny A (1993) Vacci- nation against animal retroviruses. In: Progress in Walrand F, Fumoux F, Roelants G, Parodi AL, Levy D Vaccinology. Vol 4: Veterinary Vaccines. Pandey, (1986) Incidence of bovine leukemia virus specific Hoglund, Prasad, Springler Vertag, Bertin, 87-138 antibodies in West African cattle. Int J Cancer 37, 619-621 Powers MA, Grossman D, Kidd LC, Radke K (1991) Episodic occurrence of antibodies against the bovine White KN, Nosaka T, Kanamori H, Hatanaka M, Honjo T leukemia virus rex protein during the course of infec- (1991) The nucleolar localisation signal of the HTLV- tion in sheep. J Virol65, 4959-4965 1 protein p27-rex is important for stabilisation of IL- 2 receptor. Biochem Biophys 175, 98-103 Rimsky L, Duc Dodon M, Dixon EP, Greene WC (1989) Trans-dominant inactivation of HTLV-1 and HIV-1 Whyte P, Williamson NM, Harlow E (1989) Cellular tar- gene expression by mutation of the HTLV-1 Rex gets for transformation by the adenovirus Ei A pro- transactivator. Nature (Lond) 341, 453-456 teins. Cell56, 67-75 Sagata N, Yasunuga T, Tsuzuku J, Ohishi K, Ogawa Y, Willems L, Heremans H, Chen G et al (1990) Coopera- lkawa Y (1985) Complete nucleotide sequence of tion between bovine leukaemia virus transactivator the genome of bovine leukemia virus: its evolution- protein and Ha-ras oncogene product in cellular ary relationship to other retroviruses. Proc Natl Acad transformation. EMBO J 9, 1577-1581 Sci USA 82. 677-681 Willems L, Grimonpont C, Heremans H et al (1992a) Schwartz I, Bensaid A, Polack B, Perrin B, Berthelemy Mutations in the bovine leukemia virus tax protein M, Levy D (1994) In vivo leukocyte tropism of bovine can abrogate the LTR directed transactivating activ- leukemia virus in sheep and cattle. J Virol 68, 4589- ity without concomitant loss of transforming potential. 4596 Proc Natl Acad Sci USA 89, 3957-3961 Slavikova K, Kettmann R, Reinerova M, Zajac V, Portetell Willems L, Kettmann R, Chen Get al (1992b) A cyclic D, Burn A (1987) Provirus integration of bovine AMP responsive DNA binding protein (CREB2) is a leukemia virus into DNA of infected human myeloma cellular transactivator of the bovine leukemia virus cells. Neoplasma 34, 653-657 long terminal repeat. J Virol66, 766-772 Temin HM (1993) A proposal or a new approach to a Willems L, Kettmann R, Dequied F et al (1993) In vivo preventive vaccine against human immunodeficiency infection of sheep by bovine leukemia virus mutants. virus type 1. Proc Natl Acad Sci USA 90, 4419-4420 J Virol67, 4078-4085 Valikhov AF, Burba LG, Shishkov VP (1984) Virological Xu A, Van Eijk MJT, Park C, Lewin HA (1993) Poly- examination of semen, milk and blood in BLV infected morphism in BoLA-DRB3 exon 2 correlates with cattle. In: Sth Int Symp on Bovine Leukemia Virus, resistance to persistant lymphocytosis caused by Tubingen (OC Straub, ed) Martinus Niihoff, The Hague bovine leukemia virus. J Immunol 151, 6977-6985 Van den Broecke A, Cleuter Y, Chen G et al (1988) Yoshinaka Y, Katoh I, Copelan, TD, Smythers GW, Oros- Even transcriptionally competent proviruses are silent zlan S (1986) Bovine leukemia virus protease: purifi- in bovine leukemia virus induced sheep tumor cells. cation. chemical analysis, and in vitro processing of Proc Natl Acad Sci USA 85, 9263-9267 gag precursor polyproteins. J Viral 57,826-832