Identification of the neural component of torque during manually-applied spasticity assessments in children with cerebral palsy.

Clinical assessment of spasticity is compromised by the difficulty to distinguish neural from non-neural components of increased joint torque. Quantifying the contributions of each of these components is crucial to optimize the selection of anti-spasticity treatments such as botulinum toxin (BTX). The aim of this study was to compare different biomechanical parameters that quantify the neural contribution to ankle joint torque measured during manually-applied passive stretches to the gastrocsoleus in children with spastic cerebral palsy (CP). The gastrocsoleus of 53 children with CP (10.9 ± 3.7 y; females n = 14; bilateral/unilateral involvement n = 28/25; Gross Motor Functional Classification Score I-IV) and 10 age-matched typically developing (TD) children were assessed using a manually-applied, instrumented spasticity assessment. Joint angle characteristics, root mean square electromyography and joint torque were simultaneously recorded during passive stretches at increasing velocities. From the CP cohort, 10 muscles were re-assessed for between-session reliability and 19 muscles were re-assessed 6 weeks post-BTX. A parameter related to mechanical work, containing both neural and non-neural components, was compared to newly developed parameters that were based on the modeling of passive stiffness and viscosity. The difference between modeled and measured response provided a quantification of the neural component. Both types of parameters were reliable (ICC > 0.95) and distinguished TD from spastic muscles (p < 0.001). However, only the newly developed parameters significantly decreased post-BTX (p = 0.012). Identifying the neural and non-neural contributions to increased joint torque allows for the development of individually tailored tone management.

Wetland management to reduce Baltic Sea eutrophication.

Seven regions with coastal eutrophication problems in the Baltic Sea, including the Kattegat, constitute the BERNET project (Baltic Eutrophication Regional Network). To counteract eutrophication and associated severe biological conditions the countries around this large brackish water body must all cooperate. The regions are characterized by large differences in land use, e.g. agricultural intensity, and losses of retention capacity in the catchments due to wetland reclamation. Initially it has been necessary to identify nutrient sources--especially nitrogen--and technical, economical and even administrative obstacles to initiate eutrophication management measures. Nitrogen retention in different types of wetlands in the Baltic Sea Region has been analysed. The wetlands generally have a positive effect on reduced nitrogen transport to aquatic environments and it is generally accepted that measures leading to decreased losses of nutrients to the aquatic environment must be combined with measures leading to increased retention of nutrients in catchments. Data analysed in the BERNET project show that the potential for such a measure is large. Therefore, conservation and restoration initiatives for wetlands is an essential part of the work in the BERNET project. Wetlands have been drained or totally eliminated due to intensive agriculture in some regions while large scale rehabilitation of wetlands occurs in regions with less intensive agriculture. Based on land use data from the seven regions, the working group for wetland management within the BERNET project has identified the possible use of wetlands as building blocks as a contribution to the management of the Baltic Sea eutrophication. Several recommendations are presented on the wise use of existing and constructed wetlands for water quality management in relation to non-point nutrient pollution.

Thoracoscopy as a nonpharmacotherapeutic research modification for limiting postoperative chest pain.

Diminished tissue injury and shortened clinical recovery are benefits of using an endoscopic approach for patients needing operative procedure. In the course of developing an experimental model requiring procurement of topographically precise lung biopsy specimens, we sought to apply thoracoscopy as a research alternative to thoracotomy. In addition, we investigated the influence of thoracoscopy on postprocedure recovery practices using rabbits divided into four treatment groups. Rabbit groups 1 and 2 underwent thoracoscopy and lung biopsy while maintained by one-lung anesthesia. Additionally, group 2 had ketoprofen and bupivacaine HCl analgesics injected for treatment during postprocedure recovery. These two groups were compared to control rabbits in groups 3 and 4, which underwent inhalant anesthesia without thoracoscopy. Control group 3 also received the injection analgesic combination. During recovery, rabbit behavior was systematically assessed for evidence of pain. No behavior considered indicative of pain needing intervention was observed regardless of treatment group. Limited changes in plasma corticosterone, catecholamines, and prostaglandin E2 levels measured during recovery were difficult to associate with any treatment. Unexpectedly, significantly different mean corticosterone and catecholamines levels were detected in rabbits given the injection analgesic combination in the absence of thoracoscopic procedure, as compared to other treatment groups. The results highlight the importance of awareness that analgesic drug administration has the potential to alter homeostasis and affect interpretation of some study findings by its own guise. Correlation of the mean pain study results with plasma biochemical data supports preferential use of thoracoscopy as a refinement for limiting postprocedural pain in research models.

Improved typing procedure for the polymorphic single-copy RLA-DQA gene of the rabbit reveals a new allele.

The DQA gene of the rabbit major histocompatibility complex (MHC, RLA) is highly polymorphic and, in contrast to those reported for other mammalian species, is present as a single copy. These properties allow use of this gene in a method to type the class II locus of RLA by a combination of single-stranded conformational polymorphism (SSCP) and heteroduplex (HD) analysis. Familial segregation of RLA-DQA was shown and RLA class II types for rabbits of unknown pedigree were determined using migration patterns of amplified genomic DNA. Typing results were confirmed in experiments where unknown samples were mixed with products from rabbits of RLA types defined by sequence analysis. These analyses detected an RLA-DQA allele in addition to the five previously described; this new allele is designated RLA-DQA-F.

HTLV-1 cell lines differ in constitutively activated signaling pathways that can be altered by cytokine exposure.

Examination of signaling pathways used by HTLV-1-infected rabbit cell lines revealed differences between one, RH/K30, that mediates asymptomatic infection and another, RH/K34, that causes lethal experimental leukemia. Both lines are IL-2 independent; RH/K30 produces IL-4 while RH/K34 produces IL-10. Examination of the Jak/STAT (Janus kinase/signal transducer and activator of transcription) activation of the lines revealed constitutive phosphorylation of Jak1 in both STAT6 phosphorylation, not previously reported for HTLV-1 cells, was observed in RH/K30; STAT1 and STAT3 were phosphorylated in RH/K34. Treatment with cytokines altered the activation of the STAT proteins: IL-2 induced STAT5 phosphorylation in both lines. Supernatant from RH/K34 or IL-10 induced STAT3 phosphorylation in RH/K30 cells. Supernatant from RH/K30 or IL-4 induced STAT6 phosphorylation in RH/K34 cells, which could be reversed with a Jak kinase inhibitor--AG-490.

Passage of human T-cell leukemia virus type-1 during progression to cutaneous T-cell lymphoma results in myelopathic disease in an HTLV-1 infection model.

Studies comparing functional differences in human T-cell leukemia virus type 1 (HTLV-1) clones that mediate distinct outcomes in experimentally infected rabbits, resulted in a dermatopathic smoldering adult T-cell leukemia/lymphoma following chronic infection with HTLV-1 strain RH/K34. During the 3.5 years' follow-up, HTLV-1 skin disease progressed to cutaneous T-cell lymphoma. When infection was passed to several naive rabbits, progressive paraparesis due to myelopathic neurodegeneration, analogous to HTLV-associated myelopathy, resulted in one of 4 transfusion recipients. Similar proviral loads were detected in the two diseases, regardless of stage of progression or tissue compartment of infection. Complete proviral sequences obtained from the donor and affected recipient aligned identically with each other and with the inoculated virus clone. Existence of disparate pathogenic outcomes following infectious transmission further extends the analogy of using rabbits to model human infection and disease. Although the experimental outcomes shown are limited by numbers of animals affected, they mimic the infrequency of HTLV-1 disease and authenticate epidemiological evidence of virus sequence stability regardless of disease phenotype. The findings suggest that further investigation of a possible role for HTLV-1 in some forms of cutaneous T-cell lymphoma is warranted.

Evidence for humoral and cellular reactivity against keratin and thyroglobulin in HTLV-I infected rabbits.

Human T cell leukemia virus type I (HTLV-I) infection was initially associated with T cell leukemia and a progressive neurologic disease but has since been linked to an increasing number of autoimmune disorders, including Sjogren's syndrome, uveitis, and polyarthritis. A survey of serum samples from a rabbit model of HTLV-I infection revealed that all had antibodies against keratin and thyroglobulin. Sera from several infected rabbits also reacted with collagen, while antibody reactions with other autoantigens tested, including DNA, were rare and sporadic. In addition to antibodies, cellular reactivity to keratin, but not thyroglobulin, was demonstrated by cellular proliferation in presence of IL-2 and keratin. Expanded cell cultures were positive for T cell activation markers and CD8. Association of the auto-reactivity with HTLV-I infection rather than random anti-cellular responses was supported by the fact that no antikeratin or antithyroglobulin was seen in uninfected controls, including that inoculated with uninfected lymphocytes. Finding autoantibodies in rabbits infected using naked HTLV-I DNA clones provided further assurance that infection induced the autoimmune reactions detected.

Functional characterization of a lysosomal sorting motif in the cytoplasmic tail of HLA-DObeta.

HLA-DO is an intracellular non-classical class II major histocompatibility complex molecule expressed in the endocytic pathway of B lymphocytes, which regulates the loading of antigenic peptides onto classical class II molecules such as HLA-DR. The activity of HLA-DO is mediated through its interaction with the peptide editor HLA-DM. Here, our results demonstrate that although HLA-DO is absolutely dependent on its association with DM to egress the endoplasmic reticulum, the cytoplasmic portion of its beta chain encodes a functional lysosomal sorting signal. By confocal microscopy and flow cytometry analysis, we show that reporter transmembrane molecules fused to the cytoplasmic tail of HLA-DObeta accumulated in Lamp-1(+) vesicles of transfected HeLa cells. Mutagenesis of a leucine-leucine motif abrogated lysosomal accumulation and resulted in cell surface redistribution of reporter molecules. Finally, we show that mutation of the di-leucine sequence in DObeta did not alter its lysosomal sorting when associated with DM molecules. Taken together, these results demonstrate that lysosomal expression of the DO-DM complex is mediated primarily by the tyrosine-based motif of HLA-DM and suggest that the DObeta-encoded motif is involved in the fine-tuning of the intracellular sorting.

Nucleotide sequence analyses of partial envgp46 gene of human T-lymphotropic virus type I from inhabitants of Fujian Province in Southeast China.

Partial sequences from the env(gp46) gene of two human T-lymphotropic virus type I (HTLV-I) isolates (LIN and WEN) obtained from inhabitants of Fujian Province in southeast China were analyzed. A phylogenetic tree was constructed from these sequence data and those of other known HTLV-I isolates from all over the world. Comparisons of the LIN and WEN nucleotide sequences with other HTLV-I isolates showed diversity ranging from 0.73 to 7.00% for LIN and from 0.87 to 7.00% for WEN. Sequences of isolates LIN, WEN, MT-2, TSP1, and CH were most closely related, and the phylogenetic tree showed that all belong to the widespread subtype A of the cosmopolitan group. These preliminary data indicate that HTLV-I isolates from Fujian Province, China are closely related to HTLV-I strains from Japan and the Caribbean.

Isolation and mapping of the rabbit DM genes.

Proper peptide presentation by major histocompatibility complex (MHC)-encoded class II antigens is dependent on the products of the MHC DM loci. We identified the rabbit orthologues (RLA-DMA and -DMB) of human HLA-DMA and -DMB and found that they have 76.9% and 78.8% identity with HLA-DMA and -DMB, respectively. Like classical class II MHC genes, RLA-DM genes are more closely related to human HLA-DM genes than to mouse H2-DM. Among the DM family, there is a high degree of variability at the amino terminus of the DMa chains, and length variability in the cytoplasmic tails of both DMalpha and DMbeta. The rabbit DM genes are coexpressed with class II genes in lymphoid tissues, as are the DM genes of other mammals. The RLA-DM locus maps to the class II region of the rabbit MHC, and is flanked by the DP and DOB loci. Despite having some similarities to class II genes of bony fishes, the DM family represents a separate branch of the MHC class II family.

Cellular distribution of a mixed MHC class II heterodimer between DRalpha and a chimeric DObeta chain.

Human MHC class II antigens include HLA-DR, -DQ, and -DP molecules that present antigens to CD4+ T cells, as well as the non-classical molecules HLA-DM and -DO. HLA-DM promotes peptide binding to class II molecules in endocytic compartments and HLA-DO, which is physically associated with HLA-DM in B lymphocytes, regulates HLA-DM function. Antibodies specific for the DObeta chain were obtained by immunization of mice with a heterodimer consisting of a chimeric DObeta chain (DR/DObeta), containing 18 N-terminal residues of DRbeta, paired with the DRalpha chain and isolated from transfected murine fibroblasts. The specificity of this serum for the DObeta chain and the lysosomal expression of the HLA-DO protein was confirmed using mutant human B cell lines lacking DR or DO molecules. The lysosomal localization of HLA-DO in human B cells contrasts with the cell surface expression of the mixed pair in transfected murine fibroblasts and raises questions concerning the role of the putative targeting motifs in HLA-DO. Transfection of the chimeric DR/DObeta chain along with DRalpha into human epithelial HeLa cells resulted in high levels of expression of the mixed isotypic pair at the surface of transfectants as well as in lysosomes. The same pattern was observed in HeLa cells transfected with the DObeta chimera and a DRa chain lacking the cytoplasmic tail. Taken together, these results suggest that functional sorting motifs exist in the DObeta chain but that the tight compartmentalization of HLA-DO observed inside B lymphocytes is controlled by the HLA-DOalpha chain and HLA-DM.

Rabbit cells expressing human CD4 and human CCR5 are highly permissive for human immunodeficiency virus type 1 infection.

To evaluate the feasibility of using transgenic rabbits expressing CCR5 and CD4 as a small-animal model of human immunodeficiency virus type 1 (HIV) disease, we examined whether the expression of the human chemokine receptor (CCR5) and human CD4 would render a rabbit cell line (SIRC) permissive to HIV replication. Histologically, SIRC cells expressing CD4 and CCR5 formed multinucleated cells (syncytia) upon exposure to BaL, a macrophagetropic strain of HIV that uses CCR5 for cell entry. Intracellular viral capsid p24 staining showed abundant viral gene expression in BaL-infected SIRC cells expressing CD4 and CCR5. In contrast, neither SIRC cells expressing CD4 alone nor murine 3T3 cells expressing CCR5 and CD4 exhibited significant expression of p24. These stably transfected rabbit cells were also highly permissive for the production of virions upon infection by two other CCR5-dependent strains (JR-CSF and YU-2) but not by a CXCR4-dependent strain (NL4-3). The functional integrity of these virions was demonstrated by the successful infection of human peripheral blood mononuclear cells (PBMC) with viral stocks prepared from these transfected rabbit cells. Furthermore, primary rabbit PBMC were found to be permissive for production of infectious virions after circumventing the cellular entry step. These results suggest that a transgenic rabbit model for the study of HIV disease may be feasible.

Source and route of exposure influence infectivity of a molecular clone of human T cell leukemia virus type I.

Infection with human T cell leukemia virus type I (HTLV-I) is typically asymptomatic, but does result in diverse diseases ranging from adult T cell leukemia to spastic neuromyelopathy. To date, differences in HTLV-I provirus structure have not been correlated with pathogenic or asymptomatic outcome of infection. Molecular clones of HTLV-I are now available and represent a powerful tool to link virus structure to pathogenesis. Present studies to explore in vivo infectivity and pathogenicity of an HTLV-I molecular clone, K30p, have utilized the rabbit as a model system. This clone was administered to neonatal or adult rabbits by several different routes and infectivity and pathogenicity were examined. Detection of antiviral humoral immune responses, presence of provirus in tissue samples, and isolation of virus in cultures of blood lymphocytes were used to establish systemic HTLV-I infection. Intramuscular, but not nervous system, exposure to K30p HTLV-I naked DNA resulted in infection. Conversely, neural exposure to T cells that had been transfected with the K30p HTLV-I DNA consistently resulted in systemic infection. Despite detection of HTLV-I provirus in brain and spinal cord of some infected rabbits, no clinical or neuropathological changes occurred. Source and route of virus exposure played a role in infectivity, but did not influence the pathogenic outcome of HTLV-I infection.

Genes in the pX region of human T cell leukemia virus I influence Vav phosphorylation in T cells.

Human T cell leukemia virus I (HTLV-I) causes acute leukemic disease in a low percentage of infected individuals through obscure mechanisms. Our studies compare two rabbit HTLV-I-infected T cell lines: one, RH/K34, causes lethal experimental leukemia and the other, RH/K30, mediates asymptomatic infection. We show herein that the product of the protooncogene vav is constitutively Tyr-phosphorylated in RH/K34 but not in RH/K30. A role for the retrovirus in phosphorylation of Vav was assigned by transfection experiments with molecular clones of HTLV-I derived from the two lines. The HTLV-I molecular clone from RH/K30, but not that from RH/K34, down-regulates Vav phosphorylation in a Herpesvirus ateles-transformed T cell line. Use of recombinant virus clones revealed that a pX region sequence differing by two nucleotides between the two clones mediates this down-regulation. Because Vav is involved in T cell signaling and Vav phosphorylation occurs upon activation of T cells, control of the activation state of Vav by viral proteins may relate to the leukemogenic potential of certain HTLV-I-infected cells.

Experimental perinatal transmission of human immunodeficiency virus type 1 by passage of infected T cells.

Pediatric AIDS typically follows transmission of human immunodeficiency virus type 1 (HIV-1) from infected mothers to their offspring. The possibility that infected maternal-origin cells serve as a conveyance for mother-to-child HIV-1 transmission was investigated in a rabbit infection model. Administration of HIV-1-infected human T cells to pregnant rabbits was followed by evaluation of offspring, from newborn to 1.5 years of age. HIV-1 was detected in 11 of 19 vaginally delivered offspring born to mothers given infected cells during gestation. Interstitial pneumonias or lymphoid organ lesions, similar to those seen in human pediatric AIDS, occurred in some offspring. Persistence of inoculum cell (HLA)-specific gene sequences in offspring indicated that vertical transmission can be effected by T cell-associated virus. These results along with features of rabbit biology, including primate-type placentation, short gestation, and delivery of litters, suggest that the rabbit model is advantageous for studies of perinatal HIV-1 transmission.

Infectivity of chimeric human T-cell leukemia virus type I molecular clones assessed by naked DNA inoculation.

Two human T-cell leukemia virus type I (HTLV-I) molecular clones, K30p and K34p were derived from HTLV-I-infected rabbit cell lines. K30p and K34p differ by 18 bp with changes in the long terminal repeats (LTRs) as well as in the gag, pol, and rex but not tax or env gene products. Cells transfected with clone K30p were infectious in vitro and injection of the K30p transfectants or naked K30p DNA into rabbits leads to chronic infection. In contrast, K34p did not mediate infection in vitro or in vivo, although the cell line from which it was derived is fully infectious and K34p transfectants produce intact virus particles. To localize differences involved in the ability of the clones to cause infection, six chimeric HTLV-I clones were constructed by shuffling corresponding fragments containing the substitutions in the LTRs, the gag/pol region and the rex region between K30p and K34p. Cells transfected with any of the six chimeras produced virus, but higher levels of virus were produced by cells transfected with those constructs containing the K30p rex region. Virus production was transient except in cells transfected with K30p or with a chimera consisting of the entire protein coding region of K30p flanked by K34p LTRs; only the transfectants showing persistent virus production mediated in vitro infection. In vivo infection in rabbits following intramuscular DNA injection was mediated by K30p as well as by a chimera of K30p containing the K34p rex gene. Comparisons revealed that virus production was greater and appeared earlier in rabbits injected with K30p. These data suggest that several defects in the K34p clone preclude infectivity and furthermore, provide systems to explore functions of HTLV-I genes.

Cutaneous manifestations of human T cell leukemia virus type I infection in an experimental model.

Skin diseases ranging from infective dermatitis to cutaneous lymphoma have been associated with human T cell leukemia virus (HTLV) type I. A generalized exfoliative papillated dermatopathy occurred in a rabbit 20 months into a course of chronic HTLV-I infection. Biopsies revealed epidermotropic T cell infiltrates, including Sezary-like cells, that resulted in a pattern mimicking cutaneous T cell lymphoma. HTLV-I was isolated from affected skin, and virus expression was detected in cutaneous cultures. Sezary-like cells also occurred in circulation. Interleukin-2-independent lymphocyte cultures, established from blood exhibiting elevated CD8 T cell levels and CD25 expression, had polyclonal integration of provirus. The findings are similar to those in evolving adult T cell leukemia lymphoma and may represent a prelymphomatous change. The cutaneous lymphoproliferative lesion resulted from HTLV-I infection and further establishes the New Zealand White rabbit inoculated with the RH/K34 cell line as a suitable model for investigation of HTLV-I pathogenesis.

Experimental acute adult T cell leukemia-lymphoma is associated with thymic atrophy in human T cell leukemia virus type I infection.

Human T cell leukemia virus type I (HTLV-1) infection may lead to an acutely fatal adult T cell leukemia-lymphoma (ATLL), but HTLV-1-infected people usually remain asymptomatic. Why only certain HTLV-I infections lead to acute ATLL, which is characterized by leukemic infiltration of multiple organs and immune suppression, remains unknown. A readily accessible animal model in which the spectrum of consequences resulting from HTLV-I infection can be observed would greatly aid studies of this retrovirus. New Zealand White rabbits inoculated with either HTLV-1-infected CD25+ T cells or cell-free virus, were serially necropsied at different intervals after death or humane sacrifice. Tissues were preserved at necropsy or cultured in vitro and subsequently prepared for morphologic or molecular examination. Rabbits inoculated with RH/K34, a productively infected rabbit T cell line that contains a monoclonally integrated full-length HTLV-I provirus, developed acute ATLL-like biologically malignant lymphoproliferative disease with lymphocyte infiltration of viscera; lymphomas consisting primarily of monoclonal expansions of RH/K34 manifested a variety of diffuse pleomorphic histologic types. Concurrently, lymphoproliferative disease was associated with onset of thymic atrophy in the presence of rapidly increasing thymic proviral load. In contrast, rabbits given two other HTLV-1 inocula, originally derived (as was RH/K34) using the human T cell line MT-2 as virus source, also became infected but did not develop thymic atrophy or the ATLL-like disease. HTLV-1 infection, thymic atrophy, and leukemic infiltration similar to acute ATLL occurred reproducibly in a New Zealand White rabbit model independent of RH/K34 inoculum and host histocompatibility. Thymic atrophy in RH/K34-inoculated rabbits, but not in rabbits given other similar HTLV-1, was consistent with immunosuppression sufficient to prevent rejection of the inoculum. Although the short, 8-day course of the experimental ATLL precludes its having a molecular pathogenesis identical to the human condition, the systemic consequences of acute ATLL, including its association with thymic atrophy, are closely modeled.

An antibody that binds domain 1 of CD4 inhibits replication of HIV-1, but not HTLV-I, in a CD4-positive/p56lck-negative HTLV-I-transformed cell line.

mAbs that bind to the Ig CDR3-like region in D1 domain of the CD4 molecule can inhibit the HIV-1 life cycle in CD4-positive T cells and lymphoblastoid cell lines at the stage of transcription. This antiviral effect requires the integrity of the cytoplasmic tail of CD4, which acts as a signal transduction region through its association with protein tyrosine kinases such as p56Ick. Here we investigated the role of p56Ick in the cascade of molecular events that control HIV-1 transcription in cells treated with anti-CD4 mAb directed against the Ig CDR3-like region. The Ig CDR3-like region-specific mAb, 13B8-2, blocked HIV-1 production in CD4-positive/p56Ick-negative HTLV-I-producing MT2 cells superinfected by HIV-1Lai, but had no effect on HTLV-I production, although it did inhibit Tax-induced NF-kappa B translocation. These results raise the possibility that an as yet unidentified tyrosine kinase may be capable of associating with CD4 and mediating intracellular signaling.

Replication of HIV type 1 in rabbit cell lines is not limited by deficiencies in tat, rev, or long terminal repeat function.

HIV-1 infection has been documented in rabbits, but infection proceeds slowly in this species. Human and rabbit cell lines were compared in order to identify barriers to efficient HIV-1 infection of rabbit cells. A direct comparison of human and rabbit CD4 as receptor for HIV-1 indicated that the rabbit CD4 homolog did not function well even when expressed by human cells. Examination of viral RNA production indicated that the major HIV transcripts were produced in HIV-infected rabbit cells, but were present at levels significantly lower than those found for human cells. Ability of HIV-1 LTRs to direct protein expression in human and rabbit cells was compared using gene constructs with the chloramphenicol acetyltransferase (cat) gene flanked by HIV-1 LTRs. Chloramphenicol acetyltransferase protein expression was equivalent in rabbit and human cell lines transfected with the HIV-1/CAT constructs and cotransfections with the HIV-1 tat gene led to similar increases in CAT expression. Subsequent transfections with an infectious molecular HIV clone yielded approximately equal levels of HIV protein expression in rabbit and human cell lines, suggesting that major barriers to virus production in rabbit lines exist at steps prior to transcription of the viral genome. Because HTLV-I replicates with high efficiency in rabbit cells, a chimeric virus clone was constructed consisting of the 5' portion of HIV-1 through the nef coding sequence followed by the 3' HTLV-I LTR. Transfection of most rabbit cell lines with the chimera produced levels of p24gag protein higher than those transfected with the parent HIV-1 clone. By contrast, the unmodified HIV clone replicated more efficiently in all human cell lines tested.