High-yield, in vitro protein expression using a continuous-exchange, coupled transcription/ translation system.

The Rapid Translation System (RTS 500) (Roche Molecular Biochemicals) is a high-yield protein expression system that utilizes an enhanced E. coli lysate for an in vitro transcription/translation reaction. In contrast to conventional transcription/translation, this system allows protein expression to continue for more than 24 h. We demonstrated the utility of the RTS 500 by expressing different soluble and active proteins that generally pose problems in cell-based expression systems. We first expressed GFP-lunasin, a fusion protein that, because of its toxicity, has been impossible to produce in whole cells. The second protein we expressed, human interleukin-2 (IL-2), is generally difficult to produce, either as the native molecule or as a GSTfusion protein, in a soluble form in bacteria. Finally, we demonstrated the capacity of the RTS 500 to co-express proteins, by the simultaneous production of GFP and CAT in a single reaction. This new technology appears to be particularly usefulfor the convenient production of preparative amounts (100-900 microg) of proteins that are toxic or insoluble in cell-based systems.

Rescue of myeloid lineage-committed preprogenitor cells from cytomegalovirus-infected bone marrow stroma.

The effect of murine cytomegalovirus on myelopoiesis was studied in long-term bone marrow culture to find an in vitro correlate for the lethal virus interference with bone marrow reconstitution (W. Mutter, M. J. Reddehase, F. W. Busch, H.-J. Bühring, and U. H. Koszinowski, J. Exp. Med. 167:1645-1658, 1988). The in vitro generation of granulocyte-monocyte progenitors (CFU-GM) discontinued after infection of the stromal cell layer, whereas the proliferation and differentiation of CFU-GM to granulocyte-monocyte colonies remained unaffected. A protocol was established to probe the functional integrity of earlier hematopoietic cells. Pre-CFU-GM (the progenitors of the CFU-GM) could be recovered from an infected bone marrow donor culture by transfer onto an inductive recipient stromal cell layer. Thus, at least in vitro, infection of bone marrow stroma appears to be the only cause of the defect in myelopoiesis.

Site-restricted persistent cytomegalovirus infection after selective long-term depletion of CD4+ T lymphocytes.

We have established a murine model system for exploring the ability of a CD4 subset-deficient host to cope with cytomegalovirus infection, and reported three findings. First, an antiviral response of the CD8 subset of T lymphocytes could be not only initiated but also maintained for a long period of time despite a continued absence of the CD4 subset, whereas the production of antiviral antibody proved strictly dependent upon help provided by the CD4 subset. Second, no function in the defense against infection could be ascribed as yet to CD4-CD8- T lymphocytes, which were seen to accumulate to a new subset as a result of depletion of the CD4 subset. This newly arising subset did not substitute for CD4+ T lymphocytes in providing help to B lymphocytes, and was also not effective in controlling the spread of virus in host tissues. As long as a function of these cells in the generation and maintenance of a CD8 subset-mediated response is not disproved, caution is indicated with concern to an autonomy of the CD8 subset. Third, even though with delay, the CD8+ effector cells raised in the CD4 subset-deficient host were able of clear vital tissues from productive infection and to restrict asymptomatic, persistent infection to acinar glandular epithelial cells in salivary gland tissue.

Failure in generating hemopoietic stem cells is the primary cause of death from cytomegalovirus disease in the immunocompromised host.

We have shown in a murine model system for cytomegalovirus (CMV) disease in the immunocompromised host that CMV infection interferes with the earliest detectable step in hemopoiesis, the generation of the stem cell CFU-S-I, and thereby prevents the autoreconstitution of bone marrow after sublethal irradiation. The antihemopoietic effect could not be ascribed to a direct infection of stem cells. The failure in hemopoiesis was prevented by adoptive transfer of antiviral CD8+ T lymphocytes and could be overcome by syngeneic bone marrow transplantation. CD8+ T lymphocytes and bone marrow cells both mediated survival, although only CD8+ T lymphocytes were able to limit virus multiplication in host tissues. We concluded that not the cytopathic effect of virus replication in host tissues, but the failure in hemopoiesis, is the primary cause of death in murine CMV disease.

Adoptive immunotherapy of murine cytomegalovirus adrenalitis in the immunocompromised host: CD4-helper-independent antiviral function of CD8-positive memory T lymphocytes derived from latently infected donors.

The ability of memory T lymphocytes derived from latently infected mice to control murine cytomegalovirus disease in the immunocompromised host was studied by adoptive transfer experiments. At a stage of pathogenesis when virus had already colonized target tissues, a therapeutic antiviral function could be ascribed to the CD8+ subset. This in vivo function was not restricted to sites in which intravenously infused lymphocytes usually are trapped or home in, such as the lungs or the spleen, respectively, but was also evident in the adrenal glands, a site to which antiviral effector cells have to specifically migrate. Specific infiltration of adrenal gland cortical tissue by donor-derived CD8+ memory T lymphocytes was demonstrated. CD4+ memory T lymphocytes had no antiviral effect by themselves and also were not required for the function of the CD8+ effector cells in this short-term immunotherapy model. These findings should help settle the debate about which subset of T lymphocytes comprises the effector cells that can directly control cytomegalovirus infection in the murine model system.

Molecular analysis of herpesviral gene products recognized by protective cytolytic T lymphocytes.

The infection of the mouse with murine cytomegalovirus (MCMV) served as a model system to understand the biology of human CMV infection. The contribution of cytolytic T lymphocytes (CTL) to the recovery from infection was studied. Protection against lethal MCMV disease could be conferred on immunodepleted hosts by adoptive transfer of lymphocytes. The antiviral effect was mediated by specifically sensitized T lymphocytes of the CD8+ subset. These cells limited viral spread, prevented tissue destruction by viral cytopathic effects, and protected from lethal disease. Transferred cells have protective therapeutic function even when the virus has already colonized host tissues. CD8+ cells do not require the contribution of CD4+ cells for in vivo function. Selective expression of immediate-early (IE) phase genes in target cells allowed the detection of the immunodominant IE antigen recognized by CTL. The major IE gene ieI encodes a non-structural viral phosphoprotein, pp89, which resides in the nucleus of infected cells where it acts as transcriptional regulator. Expression of gene ieI is under temporal control, and membrane presentation of the protein domain detected by CTL is down-regulated by MCMV early-phase products. A recombinant vaccinia virus expressing gene ieI induced immunity that protected mice against a subsequent challenge with a lethal dose of MCMV. The protective effect was entirely mediated by CD8+ T lymphocytes. Thus, an experimental vaccine expressing a single nonstructural herpesvirus protein can induce a protective cellular immune response.

CD8-positive T lymphocytes specific for murine cytomegalovirus immediate-early antigens mediate protective immunity.

We have shown in a murine model system for acute, lethal cytomegalovirus (CMV) disease in the immunocompromised natural host that control of virus multiplication in tissues, protection from virus-caused tissue destruction, and survival are mediated by virus-specific CD8+ CD4-T lymphocytes. Protection from a lethal course of disease did not result in a rapid establishment of virus latency, but led to a long-lasting, persistent state of infection. The CD8- CD4+ subset of T lymphocytes was not effective by itself in controlling murine CMV (MCMV) multiplication in tissue or essential for the protective function of the CD8+ CD4- effector cells. The antiviral efficacy of the purified CD8+ CD4- subset was not impaired by preincubation with fibroblasts that presented viral structural antigens, but was significantly reduced after depletion of effector cells specific for the nonstructural immediate-early antigens of MCMV, which are specified by the first among a multitude of viral genes expressed during MCMV replication in permissive cells. Thus, MCMV disease provides the first example of a role for nonstructural herpesvirus immediate-early antigens in protective immunity.

Stable expression of clonal specificity in murine cytomegalovirus-specific large granular lymphoblast lines propagated long-term in recombinant interleukin 2.

The somatic stability of cloned long-term cytolytic T lymphocyte lines (CTLL) specific for antigens encoded by murine cytomegalovirus (MCMV) was tracked for more than two years of continuous in vitro propagation. Clone S1 retained its original specificity for a structural (S) antigen of MCMV for about eight months in the presence of antigen and interleukin 2 (IL 2), but not in IL 2 alone. In the following months, however, in spite of the continued presence of antigen, clonal variants developed that displayed distinct patterns of target cell recognition, including loss of the original specificity and acquisition of exclusive specificity for the natural killer target cell YAC-1. On the other hand, large granular lymphoblast (LGL) lines, line IE1-IL and a series of sublines thereof, could be established that stably expressed Ld-restricted specificity for a viral nonstructural immediate-early (IE) antigen more than two years after withdrawal of antigen and feeder cells when propagated in the presence of pure recombinant human IL 2. The finding that the presence of antigen was not essential for the stability of clone IE1-derived CTLL indicates that maintenance of specificity in LGL lines is not a result of antigen-mediated selection, but reflects an intrinsic property.

In vivo application of recombinant interleukin 2 in the immunotherapy of established cytomegalovirus infection.

We have shown in a murine model system for cytomegalovirus (CMV) disease in the immunocompromised host that in vivo application of recombinant human IL-2 (rhIL-2) can enhance the antiviral effect of a limited number of CD8+T lymphocytes, not only in prophylaxis, but also in therapy, when virus has already colonized host tissues. The observed net effect of IL-2 was consistent with the assumption of daily effector population doublings. The prospects for IL-2-supported immunotherapy of established CMV infection depend upon the tissues involved in disease. It appears that the prospects for controlling established CMV adrenalitis are less promising than for a therapy of interstitial CMV pneumonia.

The role of Cu(I)-thiolate clusters during the proteolysis of Cu-thionein.

Rat liver Cu,Zn-[35S]thionein and yeast Cu-thionein were subjected to proteolysis in vitro using equilibrium dialysis. The partially copper-loaded vertebrate thionein (2-7 Cu/mol) was affected by different proteases including thermolysin, proteinase K, protease from Streptomyces griseus and lysosomal enzymes. Unlike the 2Cu-thionein the respective 7Cu-thiolate-centred metallothionein was hardly proteolytically digested. In contrast to fully copper-loaded native yeast Cu-thionein both the H2O2-oxidized and the metal-free protein were effectively cleaved in the presence of proteinase K. It is important to realize that the native Cu(I)-thiolate chromophore survives the proteolytic attack. When the copper-sulphur bonding is broken and the same amount of copper is unspecifically bound to the thionein portion, proteolysis proceeds identically with respect to the rate observed in the presence of the apoprotein. The unsuccessful proteolysis of native Cu-thionein is not attributable to a simple copper-dependent inhibition of the proteinases. It is suggested that prior to proteolysis the copper-sulphur clusters must be destroyed.