Differences between Mycobacterium-Host Cell Relationships in Latent Tuberculous Infection of Mice Ex Vivo and Mycobacterial Infection of Mouse Cells In Vitro

The search for factors that account for the reproduction and survival of mycobacteria, including vaccine strains, in host cells is the priority for studies on tuberculosis. A comparison of BCG-mycobacterial loads in granuloma cells obtained from bone marrow and spleens of mice with latent tuberculous infection and cells from mouse bone marrow and peritoneal macrophage cultures infected with the BCG vaccine in vitro has demonstrated that granuloma macrophages each normally contained a single BCG-Mycobacterium, while those acutely infected in vitro had increased mycobacterial loads and death rates. Mouse granuloma cells were observed to produce the IFNγ, IL-1α, GM-CSF, CD1d, CD25, CD31, СD35, and S100 proteins. None of these activation markers were found in mouse cell cultures infected in vitro or in intact macrophages. Lack of colocalization of lipoarabinomannan-labeled BCG-mycobacteria with the lysosomotropic LysoTracker dye in activated granuloma macrophages suggests that these macrophages were unable to destroy BCG-mycobacteria. However, activated mouse granuloma macrophages could control mycobacterial reproduction in cells both in vivo and in ex vivo culture. By contrast, a considerable increase in the number of BCG-mycobacteria was observed in mouse bone marrow and peritoneal macrophages after BCG infection in vitro, when no expression of the activation-related molecules was detected in these cells.


Introduction
At present, tuberculosis (TB) remains to be one of the world's most serious health concerns. In 2014, the WHO reported 9 million new TB cases and about 1.5 million deaths [1]. The increasing incidence of infection with antibiotic-resistant strains of Mycobacterium tuberculosis is an alarming trend of recent years [1][2][3]. This is indicated by an increasing incidence of acutely progressing forms of drug-resistant TB with severe clinical manifestations and a widespread occurrence of the pathological process in the organism [2][3][4]. In 2014, 480,000 new cases of fj with multiple drug resistance were diagnosed, of which only 48% recovered [1]. At present, there is the only anti-TB vaccine called the "Bacillus Calmette-Guérin" (BCG) prepared from an attenuated live strain of M. bovis, a pathogen in cattle. The BCG vaccine is considered to be only effective against disseminated forms of TB in children, but why it does not protect adults against pulmonary TB is not yet known [5,6].
Man normally becomes infected with M. tuberculosis by aerosol transmission. Pulmonary macrophages entrap mycobacteria by phagocytosis and destroy them in phagolysosomes using active forms of oxygen and nitrogen, lysosomal hydrolases, and toxic peptides in a low-pH medium. The proinflammatory cytokines IFN , IL-1 , and GM-CSF secreted by activated cells of different types exert regulation on immune processes and inflammation in response to invasion by pathogens, controlling the development of infection in the organism by enhancing the microbicidal activity of the cells of the immune system [4,6]. The uncontrollable replication of mycobacteria in the human organism normally results in acute TB; however, activation of the innate and adaptive immune responses in the human organism most often leads to latent, asymptomatic fj infection. According to the WHO, one-third of the world's population is a latent carrier of M. tuberculosis in chronic granulomatous inflammatory lesions largely composed of 2 Journal of Immunology Research macrophages [2,5,6]. Low BCG-mycobacterial loads in animal organs and tissues at different time points of chronic infection had previously been established by bacteriological methods in a model of latent tuberculous infection under which mice were infected with BCG-mycobacteria in vivo [7][8][9][10]. Using our original model of mouse granulomas in ex vivo culture, we have, for the first time, determined the bacterial load in macrophages, dendritic cells, and multinucleate Langhans giant cells in separate granulomas obtained from mice with latent tuberculous infection after in vivo exposure to BCG vaccine [11,12]. In some host cells, not only did BCG-mycobacteria survive, but also they were actively reproducing and formed cording colonies, cording being the indication of their virulence [12]. Interestingly, there was a difference in behavior between mycobacteria of virulent and nonvirulent strains in in vitro cultures of infected human, mouse, and cow cells [13][14][15][16][17][18]. Mycobacteria of virulent strains were actively reproducing in cells infected in vitro, while attenuated mycobacteria of the BCG vaccine and M. tuberculosis of nonvirulent strains were basically found in vacuoles before they were destroyed there within 2-7 days of observation in vitro [15]. However, there are very few comparative studies of relationships between mycobacteria of different strains and host cells in animals infected in vivo or following acute infection in vitro [19,20]. And very few are the studies researching relationships between BCGmycobacteria and host cells [11,12,19,21]. As is known, BCG vaccines can occasionally cause severe disease in children with inborn errors of immunity often referred to as BCG-osis [22,23]. Importantly, clinical observations of BCG infection (including BCG adenitis) in AIDS patients after as many as 30 years following BCG vaccination are still being discussed [6]. Therefore, understanding relationships between BCGmycobacteria and host cells both after infection in vivo and after acute infection in vitro is important for studying the development of BCG-induced anti-TB immunity, developing better BCG-based vaccines [5,6], and testing vaccine candidates in animal models [24], including mouse models of tuberculous and nontuberculous mycobacterial infections [24,25].
In the present work, we conducted a comparative study of the mycobacterial loads in granuloma cells from the bone marrow and spleens of mice with latent tuberculous infection following infection with BCG in vivo and several days of ex vivo culture and in the cultures of bone marrow cells and peritoneal macrophages obtained from intact mice and infected with BCG in vitro. As a result, we observed a considerable increase in the number of mycobacteria in the macrophages of bone marrow cell cultures and peritoneal macrophages within 120 h following BCG infection in vitro and the death of cells having increased BCG loads. Throughout 48-120 h of ex vivo culture, mouse granuloma macrophages each basically remained to contain a single BCG organism, and increased numbers of such microorganisms in some macrophages did not cause the host cells to die. Analysis of the levels of the proinflammatory cytokines IFN and IL-1 , the growth factor GM-CSF, and various cell-surface markers in BCG-containing macrophages in ex vivo and in vitro cultures suggested that although the active production of these molecules in mouse granuloma cells did not help in eliminating all mycobacteria in the host cells, it helped in restricting mycobacterial reproduction in granuloma macrophages. By contrast, a considerable increase in the number of BCG-mycobacteria was observed in those in vitro infected mouse bone marrow and peritoneal macrophages, whether alive or dead by apoptosis/necrosis, in which no active synthesis of these markers was going on.

Materials and Methods
2.1. Animals. Two-month-old BALB/c male mice were obtained from the Animal Breeding Facility of the Institute of Cytology and Genetics of the Siberian Branch of the Russian Academy of Sciences (Novosibirsk, Russia). Mice were bred and maintained under standard vivarium conditions, with water and food provided ad libitum. Animal experiments were conducted in accordance with "The Guidelines for Manipulations with Experimental Animals" issued by the Russian Ministry of Health (guideline 755). All experimental procedures were approved by the Local Ethical Committee of the Research Institute of Biochemistry (Novosibirsk, Russia).

Infection of Mice.
Mice were infected with a vaccine prepared from an attenuated live strain of M. bovis (the Bacillus Calmette-Guérin vaccine, BCG-1, the Institute of Microbiology and Epidemiology, Moscow, Russia) at a dose of 0.5 mg per mouse, which amounted to 3 × 10 6 viable BCGmycobacteria in 0.9% NaCl solution. Twenty-four mice were each infected via tail vein injection with 100 L of the suspension and four mice were each infected intraperitoneally with 200 L of the suspension.

Isolation and Ex Vivo Culture of Mouse Granulomas.
Mice were anesthetized and killed by cervical dislocation. Isolation of granulomas from the spleens and bone marrow of mice after 20 days, one month, and two months following infection was performed as previously described [11,12]. Bone marrow was flushed from femurs with RPMI 1640 medium (BioloT, St. Petersburg, Russia). The spleens collected from the animals were cut into small pieces in 5 mL of RPMI 1640 medium with 50 g/mL gentamicin. Granulomas were isolated from the organ homogenates by centrifugation at 150 ×g at room temperature. The supernatants were removed. Granulomas were washed three times in 10 mL of RPMI 1640 medium with 50 g/mL gentamicin by centrifugation. The granuloma pellets were resuspended in the complete RPMI 1640 growth medium containing 10% heat-inactivated fetal bovine serum (FBS), 2 mM glutamine, and 50 g/mL gentamicin (BioloT, St. Petersburg, Russia), placed at low-medium density to 24-well tissue culture plates (Orange Scientific, Belgium) with glass coverslips at the bottom and cultured in 0.5 mL medium for several days at +37 ∘ C in an atmosphere containing 5% CO 2 . Granulomas were isolated from mice 1 and 2 on day 20 following intraperitoneal infection; from mice 1 and 2 ÷ 5 after one month following intraperitoneal and intravenous infection, respectively; from mice 1 ÷ 16 and 21 ÷ 24 after two months following intravenous infection; and from mouse 25 after two months following intraperitoneal Journal of Immunology Research 3 infection. The mouse nomenclature used is explained [11,12]. Peritoneal macrophages were isolated from mice 1 and 2 after 20 days and mouse 25 after two months following intraperitoneal infection and cultured under the same conditions as the granuloma cells. After isolation of granulomas from the bones of mice 1 and 2 after 20 days following infection [12], the other bone marrow cells from supernatants were cultured under the same conditions as the granuloma cells.

Cell Cultures and Infection
In Vitro. Mice were anesthetized and killed by cervical dislocation. Peritoneal and bone marrow cells were isolated from four intact mice and used as separate mixtures of each cell culture. Peritoneal cells were isolated from peritoneal cavities in 7 mL of the complete RPMI 1640 growth medium. The femurs and tibias were removed and trimmed at both ends. Bone marrow was flushed out with 5 mL of RPMI 1640 medium containing 10% heat-inactivated FBS, 2 mM glutamine, and 50 g/mL gentamicin using a 22-gauge needle. Peritoneal macrophages and bone marrow cells (3 × 10 5 mononuclear cells/well) were placed in 24-well tissue culture plates with glass coverslips at the bottom and cultured without stimulation in 0.5 mL of the complete growth medium for 24 h under the same conditions as the granuloma cells. After removal of growth medium with nonadherent cells, the monolayer cultures of bone marrow and peritoneal cells were washed twice with RPMI 1640 medium to prepare them for infection. Dry BCG vaccine was diluted in RPMI 1640 medium, passed several times through a 26-gauge needle to obtain stand-alone bacterial cells, and was added to the cells in some of the plate wells in the amount of 0.04 mg/well (2.4 × 10 5 mycobacteria/well) in 200 L of the complete growth medium without antibiotics and incubated for 1 h at +37 ∘ C in an atmosphere containing 5% CO 2 . The cells were further washed twice with RPMI 1640 medium to remove extracellular bacteria and were cultured for various time periods under the same conditions as the granuloma cells. At hour 4 following BCG infection, more than 97% of peritoneal cells were found to be macrophages. At that time point, bone marrow cells were found to be precursor cells of different size, neutrophils, lymphocytes, and a few megakaryocytes. At hour 24 following BCG infection, the precursor cells differentiated into bone marrow macrophages both in the control (uninfected) cultures and in cell monolayers after 24 h of acute BCG infection in vitro. On day 3 after isolation of bone marrow cells, macrophages and some fibroblasts with only a few neutrophils were observed in these cultures. All macrophages in the cell cultures were CD14positive after immunofluorescent staining for the CD marker (BD Biosciences, USA, 557896, data not shown). The control fibroblasts were obtained through spontaneous cell migration from fragments of the splenic capsule from intact mice in the wells of 24-well plates with cover glasses at the bottom with a minimal amount of complete growth medium at +37 ∘ C in an atmosphere containing 5% CO 2 . After removing tissue fragments, the migrant fibroblasts on the cover glasses were further cultured for an optimal density in 0.5 mL medium for several weeks. Three independent experiments on obtaining and studying fibroblasts, cells in the mouse bone marrow and peritoneal cultures have been conducted.

Cell
Staining. At hours 48-120 of ex vivo culture, granuloma cells on coverslips were fixed with 4% formaldehyde solution in phosphate-buffered saline (PBS, pH 7.4) for 10 minutes at room temperature. The same procedure was applied to cells in bone marrow cell cultures and peritoneal macrophage cultures at hours 4, 24, 48, 72, 96, and 120 following infection with BCG in vitro and in control cell cultures. To visualize acid-fast mycobacteria within host cells and analyze their intracellular reproduction, the preparations were washed with PBS and stained by the Ziehl-Neelsen method, mycobacteria were counted under a microscope, and their mean number per macrophage was determined. The cells stained by the Ziehl-Neelsen method were further counterstained with 1% methylene blue.
In the experiments using LysoTracker Red DND-99, the cell preparations were incubated with 50 nM of the acidotropic dye for 5 minutes at +37 ∘ C in 5% CO 2 before fixation. The cell preparations were fixed as described above, washed with PBS, permeabilized for 2 minutes with 0.3% Triton-X100 solution, blocked in PBS containing 2% BSA, and finally incubated first with rabbit polyclonal primary antibodies to mycobacteria and then with Alexa 488conjugated goat polyclonal secondary antibodies to rabbit IgG.
Some of the fixed cell preparations were washed with PBS, blocked in PBS solution containing 2% BSA, and incubated first with rat monoclonal primary antibodies to mouse CD1d, CD25, CD31, and CD35, then with biotinconjugated goat polyclonal secondary antibody to rat IgG, and finally with horseradish peroxidase-conjugated streptavidin in diaminobenzidine solution containing 0.05% H 2 O 2 . After incubation, the cells were counterstained with 1% methyl green. The antibodies against CD35 were labeled with biotin. Fluorescent visualization of CD35 was achieved using Cy3-conjugated streptavidin.
Some of the fixed cell preparations were washed with PBS, treated for 2 minutes with 0.3% Triton-X100 solution, blocked in PBS solution containing 2% BSA, and incubated with rat monoclonal primary antibodies to mouse IFN or rabbit polyclonal primary antibodies to mouse fibronectin, type I collagen, and S100 proteins. The cell preparations were further incubated with goat polyclonal biotin-conjugated secondary antibodies to rat IgG or rabbit IgG. Specific staining of the cell preparations was visualized using horseradish peroxidase-conjugated streptavidin in diaminobenzidine solution containing 0.05% H 2 O 2 . The cells were further counterstained with 1% methyl green.
Some of the fixed cell preparations were washed with PBS, blocked in PBS solution containing 2% BSA, and incubated with APC-conjugated rat monoclonal primary antibodies to mouse CD11b. After staining for CD, the cell preparations were washed with PBS, treated for 2 minutes with 0.3% Triton-X100 solution, and incubated with rabbit polyclonal primary antibodies to mouse vimentin, fibronectin, type I collagen, or S100 proteins. Fluorescent visualization of the proteins was achieved using Alexa 488-conjugated goat polyclonal secondary antibodies to rabbit IgG.
Some of the fixed cell preparations were washed with PBS, blocked in PBS solution containing 2% BSA, and incubated with rat monoclonal primary antibodies to mouse CD31, GM-CSF, or IFN . Fluorescent visualization of the proteins was achieved using FITC-labeled goat anti-rat IgG polyclonal secondary antibody. After visualization, the cell preparations were washed with PBS and incubated with rat monoclonal primary antibodies to mouse CD86, CD35, and CD11b, respectively. The antibodies against CD86, CD35, and CD11b were labeled with PE-Cy7, biotin, and APC, respectively. Fluorescent visualization of CD35 was achieved using Cy3conjugated streptavidin.
Some of the fixed cell preparations were washed with PBS, treated for 2 minutes with 0.3% Triton-X100 solution, blocked in PBS solution containing 2% BSA, and incubated with TRITC-labeled phalloidin, to stain filamentous actin, and with rabbit polyclonal primary antibodies to mouse fibronectin or type I collagen. Fluorescent visualization of fibronectin and type I collagen was achieved using Alexa 488conjugated goat polyclonal secondary antibody to rabbit IgG.
Some of the fixed cell preparations were washed with PBS, blocked in PBS solution containing 2% BSA, and incubated with hamster monoclonal primary antibody to mouse IL-1 and rat monoclonal primary antibodies to mouse CD1d or IFN . Fluorescent visualization of the proteins was achieved using FITC-labeled goat anti-rat IgG polyclonal secondary antibodies and biotin-conjugated mouse monoclonal secondary antibodies to hamster IgG and Cy3-conjugated streptavidin.
Some of the fixed cell preparations were washed with PBS, blocked in PBS solution containing 2% BSA, and incubated with PerCP-Cy5.5-conjugated hamster monoclonal primary antibodies to mouse CD80. After staining for CD, the cell preparations were washed with PBS, treated for 2 minutes with 0.3% Triton-X100 solution, and incubated with rat monoclonal primary antibodies to mouse IFN . Fluorescent visualization of cytokine was achieved using goat polyclonal FITC-labeled secondary antibody to rat IgG.
The cell preparations were incubated with the appropriate antibodies and streptavidin for 60 minutes at room temperature. The preparations of peritoneal macrophages obtained from normal uninfected BALB/c mice and cultured and fixed in the same manner as granuloma cells were used as the control groups. Fluorescent staining was analyzed using the VECTASHIELD Mounting Medium with DAPI (4 ,6-diamidino-2-phenylindole) (Vector Laboratories, USA, H-1200). Confocal images of the cells were recorded; the preparations were washed in PBS for 20 minutes to remove the VECTASHIELD Mounting Medium and restained for acid-fast mycobacteria by the Ziehl-Neelsen method.

Microscopy. The cytological preparations were examined at the Shared Center for Microscopic Analysis of Biological
Objects of the Institute of Cytology and Genetics, SB RAS, using an Axioskop 2 plus microscope (Zeiss) and objectives with various magnifications (Zeiss), and photographed using an AxioCam HRc camera (Zeiss); the images were analyzed using the AxioVision 4.7 microscopy software (Zeiss). Cell preparations were stained with fluorescent dyes and examined under an LSM 780 laser scanning confocal microscope (Zeiss) using the LSM Image Browser and ZEN 2010 software (Zeiss). Granuloma cells were counted separately on each coverslip for each mouse in each test. In each experiment with bone marrow and peritoneal cell cultures, more than 1000 cells in 30 fields of view were analyzed at each time point.

Statistical Analysis.
Statistical data processing was performed using MS Excel 2007 (Microsoft). Differences were tested for significance using Student's t-test.

Mycobacterial Loads in Mouse Granuloma Cells at Different Time Points of Ex Vivo Culture and in Bone Marrow and Peritoneal Macrophages from Mice Infected with BCG In Vitro.
Granulomas from mouse bone marrow (BM/) and spleens (S/) were isolated on day 20 (/20 d) after infection with BCG in vivo, and splenic granulomas were isolated one month (/1 m) and two months (/2 m) following infection. All the granulomas were seeded into culture plates. The granulomas isolated after 20 days (this is how long it takes mice to develop adaptive immunity to BCG [9]), one month, and two months following infection were denoted as Gran/20 d, Gran/1 m, and Gran/2 m, respectively. Because none of the mice had been observed to have acute tuberculous infection at the time of granuloma isolation, it was concluded that these granulomas were isolated at the latent stage of BCG infection. Monolayer Journal of Immunology Research 5 cultures of cells that had migrated from each granuloma were obtained. Any of the monolayer cultures of granuloma cells that may or may not retain cell clusters in the center of granulomatous lesions will be referred to as the "granuloma" throughout. The cellular composition of each granuloma and the relationships between BCG-mycobacteria and host cells in granulomas isolated from mice after 20 days, one month, and two months following infection had been characterized previously [11,12]. Analysis of acid-fast BCG-mycobacterial loads in mouse granuloma cells throughout 48-120 h of ex vivo culture demonstrated that the mouse granuloma cells each basically contained a single BCG organism at all times following BCG infection in vivo and at any time point of ex vivo culture (Figure 1(a)).
The acid-fast BCG loads in the cultures of bone marrow cells and peritoneal macrophages were analyzed throughout 4-120 h following BCG infection in vitro (Figures 1(b)-1(c) and 2(a)-2(b)). While up to 60% of macrophage precursor cells of bone marrow each contained as few as two mycobacteria on average at hour 4 following BCG infection, we thereafter observed a decrease in the number of cells as those with a gradual increase in the occurrence of macrophages each containing more than 20 BCG-mycobacteria ( Figure 1(b)). At hour 120 following infection, a considerable number of macrophages in the bone marrow cell population had increased mycobacterial loads (Figures 1(b) and 2(a)). In parallel, we observed a decrease in the number of macrophages containing a single BCG organism (Figure 1(b)). In bone marrow cell cultures, a large number of neutrophils were observed to contain acid-fast BCG-mycobacteria at hour 4 following infection (Figure 2(a)). 36.52 ± 4.76%, 15.28 ± 4.75%, and 3.37 ± 0.43% of neutrophils were observed to contain 1 4, 5 9, and 10 20 BCG-mycobacteria, respectively. From hour 24 following BCG infection on, very few neutrophils were observed in bone marrow cell cultures, and those observed had no acid-fast mycobacteria in them. Throughout 24-120 h following BCG infection, bone marrow cell populations were largely composed of macrophages with diverse numbers of mycobacteria in them, a limited number of fibroblasts, some of which contained a single BCG organism, and a few dendritic cells without BCGmycobacteria.
In the mouse peritoneal macrophage cultures infected with BCG in vitro, increased acid-fast BCG-mycobacterial loads in a large number of cells were first observed at hour 4 following infection (Figures 1(c) and 2(b)). However, at hour 24 following infection, the number of cells as those was dramatically reduced, probably due to necrotic cell death, which had been indicated by the typical morphology of some cells in this population at hour 4 following infection. Throughout 48-96 h following infection, we observed a gradual increase in the number of viable macrophages with increased BCG loads in them and a decrease in the number of cells with one bacterium in each (Figures 1(c) and 2(b)). Some fibroblasts found in low numbers in the populations of peritoneal macrophages at hour 96 and 120 following infection each contained a single BCG organism. No BCG-mycobacteria were found in the dendritic cells identified in the peritoneal macrophage cultures.
Overall, our data suggest an active increase in the number of acid-fast mycobacteria in the macrophages of mouse bone marrow cell populations and mouse peritoneal macrophages within 4-120 h after BCG infection in vitro, while granuloma macrophages from the bone marrow and spleens of mice with latent tuberculous infection restricted BCG-mycobacterial reproduction in ex vivo culture. The cultures of peritoneal macrophages had quite a different pattern of cell death after infection with BCG in vitro (Figures 3(b) and 4). At hour 4 following infection, more than 20% of macrophages showed morphological evidence of being dead (Figure 3(b)). The dead cells (peritoneal macrophages) normally contained a large number of ingested BCG-mycobacteria. However, as the culture went on, there was a decrease in the number of dead macrophages: at hour 72 following infection, they made up 10% of the then current cell population at that point. However, at hour 96 following infection, dead cells made up 30% of the then current cell population (Figures 3(b) and 4), which was in correlation with the increase in the number of BCGmycobacteria in the peritoneal macrophages (Figures 1(c) and 2(b)). At hour 120 following infection, dead macrophages made up 50% of the then current cell population ( Figure 4). Each dead peritoneal macrophage contained more than one BCG organism in it-as many as each bone marrow macrophage did. Notably, some peritoneal and bone marrow macrophages with increasing mycobacterial load had the capability for phagocytosis and destruction of damaged cells, mainly lymphocytes, throughout 72-120 h following BCG infection in vitro. Phagosomes with engulfed lymphocytes at various stages of degradation existed side by side with large numbers of acid-fast BCG-mycobacteria in the bone marrow (Figure 2(a)) and peritoneal macrophages (Figure 2(b)).

Death of Macrophages with Increased BCG Loads in the
In mouse granulomas at all times following BCG infection in vivo and of ex vivo culture, macrophages with large  numbers of BCG-mycobacteria, including those in cording colonies [12], exhibited no morphological signs of cell death and retained their capacity for phagocytosis and disposal of damaged granuloma cells, mainly lymphocytes and platelets (Figure 3(c)). No colocalization of the acidophilic Lyso-Tracker probe and BCG-mycobacteria detected by antibodies reacting with the major mycobacterial cell wall component glycolipid lipoarabinomannan (LAM) was observed in granuloma macrophages, whether with single or multiple BCGmycobacteria in them ( Figure 5(a)). The LAM-labeled BCGmycobacteria that had avoided host killing in lysosomes and survived and reproduced within the granuloma macrophages were largely acid-fast, with the integrity of their cell walls preserved and, therefore, viable ( Figure 5(a)). Note that no colocalization of LAM-labeled BCG-mycobacteria with the acidotropic LysoTracker dye in the macrophages or neutrophils was observed in the cultures of bone marrow cells and peritoneal macrophages throughout 4-24 h following BCG infection in vitro either ( Figure 5(b)).
Thus, increased death rates were revealed, throughout 72-120 h following BCG infection in vitro in the cultures of bone marrow and peritoneal macrophages, for macrophages heavily loaded with BCG-mycobacteria, while in ex vivo cultures granuloma macrophages with large numbers of BCG-mycobacteria in them were still viable and had neither apoptotic nor necrotic morphology.

Production of IFN , IL-1 , and GM-CSF in Granuloma Cells from Mice with Latent Tuberculous Infection and in Bone Marrow and Peritoneal Macrophages after BCG Infection In
Vitro. It is well established that the cytokines IFN , IL-1, and GM-CSF are inflammatory mediators, which activate and differentiate immune cells, increasing their microbicidal potential in response to infection with intracellular pathogens and thus playing an important role in initiating/regulating inflammation and controlling infection [7,13,18,26,27]. An immunocytochemical and an immunofluorescence assay of IFN , cell-associated IL-1 , and GM-CSF demonstrated their active production in granuloma cells (macrophages, multinucleate Langhans giant cells, and fibroblasts) from the spleens of mice after 20 days, one month, and two months following infection with the BCG vaccine in vivo and after ex vivo culture for 48-96 h (see Table 1, Figures 6(a)-6(b), 7(b)-7(c) and 10(c)). Noteworthy, cytokine-producing granuloma macrophages with increased microbicidal potential contained replicating acid-fast BCG-mycobacteria too ( Figures  6(a), 7(c), and 10(c)). However, neither peritoneal nor bone marrow macrophages-whether after BCG infection in vitro or in the control group-were observed to synthesize these cytokines (see Table 1, Figures 6(b)-6(c)). Interestingly, there was one-fifth as many IFN -synthesizing cells in the mouse 14/2 m granulomas assayed at hour 120 of ex vivo culture (see Table 1, Figure 6   Thus, a considerable production of proinflammatory cytokines has been observed in the mouse granuloma macrophages that largely had a single BCG organism in each of them, while peritoneal and bone marrow macrophages with high bacterial loads following BCG infection in vitro have not been observed to induce the synthesis of such cytokines.

bell Surface Markers in Splenic Granuloma Cells from Mice with Latent Tuberculous Infection and in Bone Marrow
Macrophages after BCG Infection In Vitro. The bacterial lipidand the glycolipid-presenting molecule CD1d participates in T-cell development, NKT-cell activation, and regulation of immunity in response to invasion by mycobacteria [27][28][29]. CD1d molecules were found on the surface of most macrophages (both with and without acid-fast BCGmycobacteria in them), dendritic cells, and fibroblasts within the granulomas examined (Figures 7(a)-7(c) and 11). Peritoneal macrophages in the control groups stained most weakly for CD1d. Neither control fibroblasts obtained from the intact mice, nor the peritoneal macrophages from mice after 20 days following intraperitoneal infection with the BCG vaccine, nor the bone marrow macrophages from mice after 24-48 h following BCG infection in vitro stained for CD1d (Figures 7(a), 7(d), and 11). Few granuloma lymphocytes (11.13 ± 0.1% and 9.27 ± 1.14% in the granulomas from the spleens of the mice after one month and two months following infection with the BCG vaccine in vivo, resp.) stained for CD1d (Figure 7(a)).
bD25 is a marker of cell activation and the chain of the receptor for IL-2 participating in the development of the inflammatory response to alien antigens and activating antigen-presenting functions of cells of the immune system. IL-2R can combine with the other two chains of the receptor to form high-affinity signaling receptor complexes for IL-2 when macrophages and T cells are activated [30]. Additionally, bD25 is one of the markers of regulatory T cells [30]. The antibodies reacted with CD25 in 20-25% of macrophages, 4-5% of dendritic cells, and 6-12% of lymphocytes in the mouse granulomas (Figures 8 and 11). Note that the number of lymphocytes containing bD25 was on average higher in granulomas from the mice after two months than one month following BCG infection in vivo (Figure 11). The granuloma fibroblasts did not stain for this marker (Figure 8). CD25 was not found in peritoneal macrophages from the control mice, peritoneal macrophages from mice after 20 days following intraperitoneal infection with the BCG vaccine, or bone marrow macrophages at hour 48 following BCG infection in vitro (Figures 8 and 11). CD31, also known as PECAM-1 (Platelet Endothelial Cell Adhesion Molecule 1), is a transmembrane glycoprotein, one of the adhesion molecules in the immunoglobulin superfamily. Its primary function is the mediation of leukocyteendothelial cell adhesion and signal transduction in various inflammatory disorders [31]. CD31 was found on the surface of 17-27% of macrophages, both with and without acid-fast BCG-mycobacteria in them, and on the surface of 30-40% of dendritic cells in granulomas from all mice infected with the BCG vaccine in vivo (Figures 9 and 11). CD31 was not detected on fibroblasts from the spleens of the control mice, nor was it found on mouse granuloma fibroblasts (Figure 9(a)). The peritoneal macrophages obtained from the intact mice or the mice after 20 days following intraperitoneal infection with the BCG vaccine stained little for CD31 (Figures 9 and 11). CD35, also known as the complement receptor type 1 (CR1), participates in phagocytosis of b3b-opsonized mycobacteria and promotes microbial death in the phagosomes of host cells [32,33]. CR1 was found on the surface of 22-39% of the macrophages, both with and without acidfast BCG-mycobacteria, and 9-13% of the dendritic cells in mouse granulomas (Figures 10(a), 10(c), and 11). The peritoneal macrophages obtained from the intact mice or the mice after 20 days and after two months following intraperitoneal infection with the BCG vaccine did not stain for CD35 (Figures 10(a) and 11). CD35 was not detected on fibroblasts from the spleens of the control mice or mouse granulomas. Also, bone marrow macrophages stained little for CD35 throughout 24-48 h after BCG infection in vitro (Figures 10(c) and 11).
Thus, this study of the cell-surface markers demonstrates significant activation of granuloma cells through the expression of various receptors in the mouse granulomas with low mycobacterial loads; however, bone marrow macrophages that had been infected with BCG in vitro and had very high bacterial loads were not observed to produce any of those activation-related molecules.

Extracellular Matrix and Cytoskeletal Proteins in Splenic Granuloma Cells from Mice with Latent Tuberculous Infection.
It has been established that mycobacteria have evolved a wide range of molecules, known as adhesions, to enable binding with fibronectin, a protein of the host extracellular matrix, and type I collagen fibers and, eventually, to invade host cells [34]. The cytoskeletal filamentous proteins actin and vimentin are responsible for cell migration, attachment to the substrate, maintaining cell shape, integrity of the cytoplasm, phagocytosis of bacterial agents, and elimination of intracellular pathogens [34]. According to preliminary results of the immunocytochemical and the immunofluorescence assay, the production rates for fibronectin, type I collagen, filamentous actin, and vimentin in granuloma macrophages were not affected by mycobacterial loads, nor were those in granuloma fibroblasts from mice following infection with BCG vaccine in vivo or in peritoneal macrophages and splenic fibroblasts from the control group of intact mice (Figures 12(a)-12(e) and 13). No colocalization of BCG-mycobacteria with filamentous actin in granuloma macrophages was observed either (data not shown).
The S100 proteins are a large family of cytoplasmic and extracellular Ca 2+ -binding molecules regulating enzyme activities, the dynamics of cytoskeletal constituents, Ca 2+ homeostasis, and the inflammatory response [35]. The S100 proteins regulate intermediate filaments, microfilaments and other cytoskeletal proteins, cell morphology and motility, and the reciprocal relationships of cytoskeletal elements [36]. Mouse granuloma macrophages stained much stronger with polyclonal antibodies reactive with the S100 proteins than peritoneal macrophages from the control mice ( Figure 14).
Thus, a high expression of cytokines and other activationrelated CD markers, the unchanged synthesis of constitutive cytoskeletal and extracellular matrix proteins, was observed in granuloma cells with low mycobacterial loads from mice with latent tuberculous infection. By contrast, no expression of the cytokines or CD markers was detected in the control cultures of uninfected cells or peritoneal and bone marrow macrophages with high bacterial loads following acute BCG infection in vitro.

Discussion
In the present study, we have for the first time compared Mycobacterium-host cell relationships in individual granuloma cells from the bone marrow and spleens of mice with latent tuberculous infection and in the cells from mouse bone marrow and peritoneal cultures after BCG infection in vitro. The study of BCG loads in mouse granuloma macrophages at the latent stage of tuberculous infection demonstrated that, although there was individual variation in the number of bacteria across the cells of granulomas from different mice [12], those cells each normally contained a single BCG organism throughout 48-120 h of ex vivo culture. By contrast, a considerable increase in the number of BCG-mycobacteria was observed in the cultures of bone marrow cells and peritoneal macrophages infected with BCG in vitro throughout 4-120 h following infection. Therefore, mouse granuloma macrophages could control mycobacterial reproduction in cells, while the intact mice's macrophages infected in vitro could not do as much. In a study by Jordao and the coworkers [21], BCG-mycobacteria were killed very efficiently by mouse bone marrow-derived macrophages, but not by human monocyte-derived macrophages, in which BCG-mycobacteria actively replicated for 2-7 days after BCG infection in vitro.
As is known, the proinflammatory cytokines IFN , TNF , and IL-1 synthesized by macrophages, lymphocytes, and other cells of the immune system in animals control tuberculous infection [37][38][39]. It has been demonstrated that the action of IFN on cells infected with mycobacteria in vitro leads to activation of inducible NO synthase and necrotic/apoptotic death of infected cells [21,26,40]. In our    mice, causing increased death rates for host cells. In contrast to the cultures of cells infected with the BCG vaccine in vitro, mouse granuloma macrophages were not observed to have a burst-like increase in BCG-mycobacterial population during ex vivo culture or to die because of increased BCGmycobacterial loads. Our results are not consistent with those reported by Lee and Kornfeld [40], who had added recombinant cytokine IFN to the growth medium with mouse bone marrow macrophages infected with BCG-mycobacteria and M. tuberculosis in vitro and observed increased death rates for the cells that were heavily loaded with mycobacteria. However, our results are consistent with the conclusion made by the same authors [40] that IFN promotes survival of macrophages with a low intracellular bacterial load in them by inhibiting mycobacterial reproduction. Zhang and the coworkers [41] demonstrated that one of the mechanisms that contribute to the restriction of M. tuberculosis reproduction in host cells could be depletion of tryptophan in activated macrophages through IFN -mediated upregulation of expression of the host enzyme that degrades tryptophan in mouse cells, which reduces the amount of tryptophan required for intracellular mycobacteria to reproduce actively.
Curiously, we revealed that host cells containing large numbers of BCG-mycobacteria in the bone marrow cell cultures and peritoneal macrophage cultures had been dying out at increased rates without whatever external factor, not with, for example, recombinant IFN used by Lee and Kornfeld [40] or Jordao and the coworkers [21]. In our study, the observed pattern of the death of peritoneal macrophages infected with BCG-mycobacteria in vitro demonstrated that macrophages are able to ingest large enough numbers of mycobacteria; however, some contained too many to deal with, and so they died in the first hours following infection. The remaining macrophages, which had ingested few mycobacteria, were viable until the growing bacterial populations reached the size at which host cells could not carry on, which, again, resulted in more dead macrophages. A similar pattern of macrophage death was observed in BCGinfected bone marrow cell cultures with the difference that the infected bone marrow macrophage precursor cells each normally ingested no more than two BCG-mycobacteria, which had no adverse effect on their viability in vitro. The subsequent active reproduction of BCG-mycobacteria in the bone marrow macrophages differentiated from the precursor cells resulted in above-critical BCG-mycobacterial loads in host cells, and so those host cells died within 96-120 h of culture following BCG infection in vitro. Lee and the coworkers [42] found no morphological or biochemical evidence of the apoptotic death of host cells following in vitro infection of mouse bone marrow cells with BCG and M. tuberculosis. We, by contrast, observed dead BCGinfected peritoneal and bone marrow macrophages with nuclear chromatin condensation, nuclear fragmentation, and the formation of apoptotic vesicles with and without BCGmycobacteria, which was indicative of apoptosis. Some cells had plasma membrane disruptions and no nuclei in them, which was indicative of necrosis; however, acid-fast BCGmycobacteria were normally located in the pieces of the cytoplasm of dead cells, just as they were in the previously studied mouse granulomas [12], in which BCG-mycobacteria had occasionally been found in apoptotic bodies and in the fragments of the cell cytoplasm located in the phagosomes of macrophages or outside cells. The apoptotic death of invaded cells is perhaps the best option for the host organism, because this leads to the elimination of the pathogen, as apoptotic bodies are ingested by nearby macrophages, and prevents the extracellular spread of mycobacteria and the development of inflammatory responses in the surrounding tissues [37,[43][44][45].
In this work, we compared the production rates for cytokines and activation-related CD markers in granuloma macrophages from mouse spleens at different time points of latent tuberculous infection following infection with the BCG vaccine in vivo and in mouse bone marrow and peritoneal macrophages following infection with the BCG vaccine in vitro and after culture for several hours. As a result, we revealed a considerable production of proinflammatory cytokines in the mouse granuloma macrophages that largely had a single BCG organism in each of them, while peritoneal and bone marrow macrophages with high bacterial loads following BCG infection in vitro were not observed to induce the synthesis of such cytokines. Note that literature data on the induction of expression of proinflammatory cytokines in animal and human cells infected with pathogenic, nonpathogenic, and attenuated mycobacterial strains in vitro are quite controversial. For example, Beltan and the coworkers [46] Figure 1.

Control Mph Gran-3,
Gran-8, mouse 3/2 m Figure 14: The S100 proteins in the cells in the fragments of mouse splenic granulomas and control mouse peritoneal macrophages (Mph). The scale bars are 10 m each. In the left panel, confocal immunofluorescent localization of S100 proteins (green signal) and CD11b (red signal). Colocalization of the markers on confocal images of cells (yellow signal). Nuclei are stained by DAPI (blue signal). In the central and right panels, immunochemical localization of S100 proteins in the cells. The brown color indicates the presence of the markers in these cells. The black and red arrows indicate macrophages and a dendritic cell, respectively, with increased production of S100 proteins. Abbreviations as in Figure 1. than did infection of cells with less virulent mycobacterial strains [47]. In a work by Rajaram and the coworkers [48], human monocyte-derived macrophages infected in vitro with virulent M. tuberculosis H37Rv and attenuated BCG-mycobacteria followed a similar pattern of activation of the production of the proinflammatory cytokine IL-8. However, in a work by Brooks and the coworkers [49], BCGmycobacteria induced a stronger activation of the synthesis of the proinflammatory cytokine IL-1 in mouse bone marrow macrophages than in cells of the same type infected with M. tuberculosis H37Rv in vitro. As a result, these researchers observed a stronger reproduction of the mycobacteria of the virulent strain than BCG-mycobacteria throughout 24-168 h following infection [49]. Interestingly, the in vitro infection of mammalian cells with BCG-mycobacteria led to a rapid elimination of BCG-mycobacteria in the phagosomes of infected cells in some cases [15], while in others a considerable increase in the number of BCG-mycobacteria was observed during in vitro culture of infected cells [26,49]. It should be noted that Sanarico and the coworkers [50] also described the BCG strain-specific modulation of the bacterium-host cell interactions in human monocyte-derived dendritic cells after acute infection with BCG Japan or BCG Aventis Pasteur in in vitro culture. As is known [51], BCG Japan belongs to the "early strains" of BCG as does BCG Russia used in our study. On the other hand, BCG Japan was incapable of intracellular replication, while BCG Aventis Pasteur belonging to the "later strains" enlarged the bacterial population in dendritic cells [50]. Nevertheless, the differences in the gene profile were not observed in dendritic cells infected with various BCG strains, when the highest levels of mRNA for the proinflammatory cytokines IL-1 , IL-1 , IL-6, IL-12, and TNF were detected. Analysis of IFN gene expression did not show any significant modulation of the production of this cytokine in dendritic cells after BCG infection [50]. In our study, relationships between mycobacteria of the vaccine strain BCG and host cells were quite different in splenic granulomas from mice with latent tuberculous infection and in mouse cells acutely infected with BCG-mycobacteria in vitro. We observed, as Brooks and the coworkers did [49], a stronger reproduction of BCG-mycobacteria in the macrophages of those cultures, whose cells were not producing proinflammatory cytokines or CD markers of cell activation. Thus, our study demonstrated that the in vitro infection of mouse cells in culture and the in vivo infection of animals cause macrophages to produce different responses and entail different relationships between host cells and mycobacteria. The reasons for the inconsistent results and large variation in the intracellular growth rate of mycobacteria among different research groups are unclear; however, as reviewed by Rivero-Lezcano [52], mycobacterial strains, cell models, infection protocols, assays of mycobacterial growth, and cytokines for activation of host cells were different and so were the experimental settings. In addition, the observed differences should be taken into account whenever experimental cell models are used for testing and validation of antimycobacterial drugs and vaccines. Overall, although the differences in Mycobacterium-host cell relationships are actively studied by many researchers, more information is required for a better understanding of pathogen-host cell interactions. Therefore, as reviewed in [52], the development of more complex cell models that can involve most types of immune cells and many humoral immunological factors may help discover new antimycobacterial mechanisms. Consequently, we have developed a new ex vivo granuloma cell model from mice with latent tuberculous infection that is very similar to in vivo conditions, both for research into Mycobacterium-host cell relationships in the short term and for testing new vaccines and antimycobacterial drugs in a long term.
We have observed induction of expression of the activation markers CD1d, CD25, CD31, and CD35 on mouse granuloma macrophages at different time points of latent tuberculous infection, and not on bone marrow macrophages after infection with the BCG vaccine in vitro. As is known [29,53], CD1d is an antigen-presenting receptor of nonprotein, lipid, and glycolipid molecules to a variety of T cells. LAM has been identified as a CD1 antigen in mycobacteria [29]. It is possible that some CD1d-positive mouse granuloma lymphocytes were the so-called invariant natural killer T cells, which respond very rapidly to mycobacterial infection [29]. Activation of CD25 on mouse granuloma macrophages demonstrated an increased responsiveness of the macrophages to the powerful immunoregulatory lymphokine IL-2, which stimulates macrophages and other cells of the immune system for the treatment of infectious diseases [30]. It is possible that the lymphocytes with CD25 in mouse granulomas are in the class of regulatory T cells involved in suppression of activation of the microbicidal functions of macrophages and dendritic cells with the development of immune tolerance [54], so much so that CD25positive lymphocytes were observed to increase in number as chronic tuberculous infection in mice was progressing. The increased number of CD31 and CD35 receptors on granuloma macrophages was indicative of activation of the proinflammatory response of cells to mycobacterial infection in mice. Elevated S100 protein concentrations in mouse granuloma cells were probably due to activation of the production of calgranulins, because these three proteins, S100A8, S100A9, and S100A12, are inducible in macrophages and are largely expressed by activated macrophages as important mediators of inflammation and microbial infections, suggesting a key role for these proteins in the innate immune response [36]. Nevertheless, although the cells of the immune system in the mice were activated, BCG-mycobacterial reproduction in mouse granuloma macrophages was observed to go on.
As is known [34], mycobacteria possess a large number of bacterial proteins that operate as bacterial adhesions for binding with proteins of the extracellular matrix, followed by infection of host cells through activation of phagocytosisrelated processes. Preliminary results suggested that there was no difference in the production of the extracellular matrix proteins fibronectin and type I collagen between mouse granuloma cells, whether containing BCG-mycobacteria or not, and the control mouse macrophages. Furthermore, no differences were observed in the expression of the cytoskeletal proteins vimentin and filamentous actin. Studies of the proteins that are involved in cell-to-cell interactions and phagocytosis of bacteria in the cells operating within inflammatory foci such as granulomas will go on, because they are important for understanding the mechanisms underlying mycobacterial persistence in host cells in the settings of latent tuberculous infection in animals and man.
Consequently, the search for factors that influence the reproduction and survival of mycobacteria, including vaccine strains, in host cells and the study of relationships between cells and their intracellular pathogens, both following acute infection in in vitro culture and in foci of latent tuberculous infection, are the priority in investigating the role of the cells of the immune system in the protection of man and animals against mycobacterial infections caused by pathogenic or nonpathogenic mycobacteria.

Conclusions
BCG has been used as a vaccine for immunization of many people worldwide; however, the results are varying. Considering the situation, research into the host cell response to BCG-mycobacteria during acute and latent chronic infection of animal cells is very important for developing a more rational approach to the improvement of the BCG vaccine. In the present study, we compared Mycobacterium-host cell relationships in individual granuloma cells from mice with latent tuberculous infection and cells from mouse bone marrow and peritoneal cultures infected with BCG in vitro. This comparison showed different specific cellular responses to latent chronic and acute BCG infection in vitro. We have demonstrated that mouse granuloma macrophages with an increased production of proinflammatory cytokines, growth factors, and CD receptors for cell activation cannot eliminate intracellular BCG-mycobacteria. Nevertheless, these macrophages can control mycobacterial reproduction in host cells both in vivo and in ex vivo culture, while those not activated after acute BCG infection in vitro cannot. We expect that our results will add to the understanding of BCG-Mycobacterium-host cell interactions and may facilitate the design and improvement of vaccines against tuberculosis.