As a paradigm for chronic infectious diseases, tuberculosis exhibits a variety of clinical presentations, ranging from primary pulmonary tuberculosis to reactivation tuberculosis and cavitary disease. To date, the animal models used in evaluating chemotherapy of tuberculosis have been high-dose intravenous models that mimic the disseminated forms of the disease. In the present study, we have used a low-dose aerosol exposure model which we feel better reflects newly diagnosed tuberculosis in patients converting to tuberculin positivity. As appropriate examples of chemotherapy, four rifamycins (rifampin, rifabutin, rifapentine, and KRM-1648) were tested, first in an in vitro murine macrophage model and then in the low-dose aerosol infection model, for their activity against Mycobacterium tuberculosis. In both models, KRM-1648 had the highest level of activity of the four compounds. In the infected-lung model, rifabutin, rifapentine, and KRM-1648 all had sterilizing activity when given orally at 5 mg/kg of body weight per day. When given at 2.5 mg/kg/day, KRM-1648 had the highest level of activity of the four drugs, reducing the bacterial load by 2.7 logs over 35 days of therapy.

The capacity of metronidazole to inhibit the growth ofMycobacterium tuberculosis was tested in in vitro and in vivo mouse models. In vitro addition of metronidazole to cultures of infected bone marrow-derived macrophages had no effect, nor did it increase the reduction in bacterial load due to isoniazid. In vivo, metronidazole did not reduce bacterial numbers in the lungs of aerosol-infected mice during the active stage of the disease, during a phase of containment, or after prolonged isoniazid therapy (Cornell model). A small but significant reduction was seen if metronidazole therapy was given during an established chronic disease state 100 days after aerosol administration. These data indicate that under most conditions M. tuberculosis organisms are not in a metabolic state in which they are susceptible to the action of metronidazole and, hence, that this drug would be of limited clinical value.

The results of this study show that clinical isolates ofMycobacterium avium fail into two categories in terms of their capacity to grow within murine bone marrow-derived macrophage cultures: those that grow progressively and those that are incapable of growing within such cells. Members of the first category were invariably of the smooth-transparent colonial type, while most of the second were of the smooth-domed type. In addition, this paper shows that although all isolates induced tumor necrosis factor (TNF) secretion by host cells to some extent, this production was always delayed in isolates that subsequently grew well in the host cells. This observation, coupled with the demonstration that the growth of the latter isolates was inhibited by the exogenous addition of TNF, leads us to hypothesize that the ability ofa given isolate to somehow avoid host macrophage TNF production early during the course of the infection is a key factor in the pathogenesis of the disease.Infections caused by the Mycobacterium avium complex are the most common causes of bacteremia and disseminated multiorgan bacterial disease in patients with AIDS. M. avium infection is seen clinically in about 25% of patients and is detected upon autopsy in over 50%. To date, treatment of these infections is very difficult because of the high levels of resistance of most M. avium isolates to antimycobacterial drugs (4, 11, 13-15, 17, 25).Isolates of M. avium occur in at least three forms: flat or smooth transparent (SmT), raised or smooth domed (SmD), and rough. All can be observed in cultures of clinical isolates, but SmT is the most prevalent form and is the form most commonly associated with active, disseminated disease (18). There is also an association between colonial morphology and the virulence of the organism, in that recent reports have shown that SmT but not SmD isolates grow well in human monocytes (6,19,24) and that the two forms elicit different cytokine patterns from the infected host cells (19).In the latter report (19), differential cytokine release was reported only for cloned variants derived from a single isolate, so it was unclear whether this represented a general trend. In view of this, in the present study, we examined the ability of a panel of clinical isolates of M. avium to grow in bone marrow-derived murine macrophages. In addition, the cell culture supernatants were harvested at various times and tested for the presence of various cytokines. The results of this study show that while the production of most cytokines tested did not follow any definite trend, there was a marked distinction between the early production of tumor necrosis factor (TNF) by macrophages infected with progressively growing isolates and that by macrophages infected with isolates incapable of growing in these host cells. Our further observation that addition of exogenous TNF was protective regardless of the growth status of the organism leads us to hypothesize that the capacity of a given isolate to * Corresponding author. somehow avoid the product...

Even though the macrophage is the host cell for the intracellular bacterial parasite Mycobacterium avium, macrophages have undergone only limited evaluation as models for determining the capacities of antimycobacterial drugs to inhibit the growth of M. avium within this relevant intracellular environment. In the present study, we demonstrated that a panel of M. avium isolates could actively infect homogeneous monolayers of murine bone marrow-derived macrophages. A number of established and experimental antimycobacterial drugs were then added to these cultures at a range of concentrations, and their effects on the numbers of surviving bacilli were determined 8 days later. By plotting such numbers versus drug concentrations it was then possible to clearly distinguish between compounds with bactericidal activity (such as rifabutin and PD 125354) and those with bacteriostatic effects (such as clarithromycin), even though several of these compounds had very similar MICs. In addition, an estimate of the potential therapeutic efficiency of each drug could be made by determining the concentration needed to destroy an arbitrary percentage of the inoculum (in this case, the bactericidal concentration destroying 99%o of the inoculum). Such values were considerably in excess of the MICs and may more realistically reflect the concentrations in serum required to electively reduce the bacterial burden in vivo.

To date, most new vaccines against Mycobacterium tuberculosis, including new recombinant versions of the current BCG vaccine, have usually been screened against the laboratory strains H37Rv or Erdman. In this study we took advantage of our recent work in characterizing an increasingly large panel of newly emerging clinical isolates [from the United States or from the Western Cape region of South Africa], to determine to what extent vaccines would protect against these [mostly high virulence] strains. We show here that both BCG Pasteur and recombinant BCG Aeras-422 [used here as a good example of the new generation BCG vaccines] protected well in both mouse and guinea pig low dose aerosol infection models against the majority of clinical isolates tested. However, Aeras-422 was not effective in a long term survival assay compared to BCG Pasteur. Protection was very strongly expressed against all of the Western Cape strains tested, reinforcing our viewpoint that any attempt at boosting BCG would be very difficult to achieve statistically. This observation is discussed in the context of the growing argument made by others that the failure of a recent vaccine trial disqualifies the further use of animal models to predict vaccine efficacy. This viewpoint is in our opinion completely erroneous, and that it is the fitness of prevalent strains in the trial site area that is the centrally important factor, an issue that is not being addressed by the field.