| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
National Hansen’s Disease Programs, Laboratory Research Branch, Louisiana State University School of Veterinary Medicine, Baton Rouge, LA 70803
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) for its maintenance. We have developed an in vitro model to study the fate of Mycobacterium leprae in a LL lesion, with and without immunotherapeutic intervention. Target cells, consisting of granuloma M harvested from the footpads of M. leprae-infected athymic nu/nu mice, were cocultured with normal or IFN-
-activated (ACT) effector M. The bacilli were recovered and assessed for viability by radiorespirometry. M. leprae recovered from target M possessed high metabolic activity, indicating a viable state in this uncultivable organism. M. leprae recovered from target M incubated with normal effector M exhibited significantly higher metabolism. In contrast, bacilli recovered from target M cocultured with ACT effector M displayed a markedly decreased metabolic activity. Inhibition by ACT M required an E:T ratio of at least 5:1, a coculture incubation period of 3–5 days, and the production of reactive nitrogen intermediates, but not reactive oxygen intermediates. Neither IFN- nor TNF-
were required during the cocultivation period. However, cell-to-cell contact between the target and effector M was necessary for augmentation of M. leprae metabolism by normal effector M as well as for inhibition of M. leprae by ACT effector M. Conventional fluorescence microscopy and confocal fluorescence microscopy revealed that the bacilli from the target M were acquired by the effector M. Thus, the state of M infiltrating the granuloma may markedly affect the viability of M. leprae residing in M in the lepromatous lesion. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Clinical leprosy is classified as an immunohistopathological spectrum (2). In lepromatous leprosy (LL), 3 the immune response exhibits a Th2-type cytokine profile with a poorly developed cell-mediated immune (CMI) response and numerous acid-fast bacilli (AFB). At the opposite end of the spectrum in tuberculoid leprosy (TT), the lesions display characteristics of well-developed CMI, possess few AFB, and express a Th1-type cytokine profile. Between these two polar forms of leprosy is the highly unstable borderline area of the spectrum, which exhibits characteristics ranging from borderline tuberculoid to mid-borderline to borderline lepromatous disease. A change in the host’s immune response can result in upgrading or downgrading in the spectrum or in acute reactional episodes.
There is no doubt that T cells play the crucial and deciding role in orchestrating the adaptive immune response in leprosy. MHC-restricted CD4+ and CD8+ T cells (3), as well as CD1-restricted T cells (4, 5), and 
T cells (6) are all involved. In addition, experimental evidence verifying the importance of T cells in the immunoregulation of leprosy has been provided by infection of the athymic nu/nu mouse. M. leprae inoculated into the footpads grow to enormous numbers, yielding up to 1010 organisms in a lepromatous-type lesion (7, 8), in contrast to the limited growth and pathology seen in the footpads of immunocompetent mice (9).
The mononuclear phagocyte, or macrophage (M), is the preferred host cell for M. leprae. Normal M cannot kill M. leprae (10), and if no M. leprae-specific T cell immune response is generated, as in LL and the nu/nu mouse footpad, the M remain unactivated, and growth of the bacilli proceeds essentially unchecked. Implicit in this response is that new M must migrate into the lesion to accommodate the increasing bacterial load. How does the new M acquire M. leprae during growth of the bacilli? Does overwhelming bacterial multiplication cause the M to burst and release the intracellular bacilli, which are then phagocytosed by new M, or does the newly arrived M play a more active role in acquiring bacilli from the old M?
Toward the tuberculoid end of the spectrum, M function as potent effector cells of resistance and are responsible for killing and eliminating M. leprae. T cells generate the M activation factor, IFN-, and if M are activated (ACT) with IFN- before infection with M. leprae, they can efficiently kill the bacilli (10). However, there is evidence that M. leprae-infected M, especially if heavily infected for a prolonged period of time, are refractory to activation by IFN- (11, 12, 13). This implies that killing of M. leprae in infected M is likely accomplished by new ACT M that migrate into the lesion in response to various chemotactic stimuli. How does the new ACT M acquire and kill M. leprae? There is ample evidence that M. leprae-specific CTLs are generated at the tuberculoid end of the spectrum, and that these T cells lyse M. leprae-infected M and Schwann cells (14, 15, 16). This is most likely the primary mechanism of cell turnover in TT. However, we wondered whether or not ACT M play a role in cell turnover; there are precedents that suggest that they may. First, numerous tumor models have shown that M can act as killer cells and attack abnormal cells (17, 18, 19, 20, 21, 22). Second, immunotherapy with Th1-type cytokines in LL patients demonstrated upgrading of clinical disease in the absence of specific T cell-mediated immunity (23, 24).
Therefore, as part of our efforts to develop an in vitro model for the leprosy granuloma, we investigated the fate of M. leprae from infected target M upon coculture with new M. We examined whether new M could acquire the bacilli from infected target M, if the viability of the bacilli was modified, and what M effector mechanisms were involved. We provide evidence that effector M, both normal and ACT, acquired the bacilli from the target M in a contact-dependent manner. Coculture of infected target M with normal effector M augmented M. leprae metabolic activity, whereas coculture with ACT effector M decreased M. leprae metabolism via a reactive nitrogen intermediate (RNI)-dependent pathway.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 (B6) mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Enhanced green fluorescent protein (EGFP)-expressing transgenic mice were originally obtained from The Jackson Laboratory and were bred in our vivarium. Inducible NO synthase knockout (NOS2–/–) mice (The Jackson Laboratory) and mice deficient in the phagocyte oxidase gp91 subunit (phox91–/–) (The Jackson Laboratory), as well as athymic nu/nu mice (The Jackson Laboratory), all on a B6 background, were housed under aseptic conditions in microisolators (MicroIsolator Mouse VCL Exhaust Rack Housing System; Lab Products, Seaford, DE).
Cultivation and maintenance of M. leprae
M. leprae are maintained in the footpads of nu/nu mice by programmed passage to assure a weekly supply of fresh, highly viable bacilli (25).
Cell culture
Bone marrow-derived M.
Bone marrow M were cultured as previously described (26). Briefly, cells were harvested from both femurs and seeded onto 13-mm thermanox coverslips (Nalge Nunc International, Naperville, IL) in 24-well tissue culture plates (Corning, Corning, NY) in culture medium (DMEM containing HEPES and sodium bicarbonate (Life Technologies Invitrogen, Carlsbad, CA), 10% FBS (HyClone, Logan, UT), 2 mM L-glutamine (Life Technologies Invitrogen), and 50 µg/ml gentamicin (Sigma-Aldrich, St. Louis, MO)) supplemented with 5 µg/ml M-CSF (R&D Systems, Minneapolis, MN). After 6 days of incubation at 37°C, nonadherent cells were removed with vigorous washing in PBS (Irvine Scientific, Santa Ana, CA), and the adherent M monolayer was incubated in culture medium without M-CSF for an additional day.
Resident peritoneal M.
Resident peritoneal cells were harvested by lavage and cultured at 2 x 106 cells per well on coverslips in culture medium overnight at 37°C, and the adherent M population was purified by vigorous washing to remove nonadherent cells.
Footpad granuloma M.
Granuloma M were harvested from the footpads of athymic nu/nu mice, infected 9–14 mo previously with 1 x 108 viable M. leprae using a modification of the method of Sibley and Krahenbuhl (11). Granuloma tissue was aseptically removed from each footpad and carefully sliced into small fragments with scalpels. The tissue fragments were digested in 2 ml of RPMI 1640 (Life Technologies Invitrogen) containing 20% FBS and 0.7 mg/ml collagenase XI-S (Sigma-Aldrich) and 30 µg/ml DNase 1 (Sigma-Aldrich) by end-over-end rotation for 30 min at 37°C. The tissue digests were incubated on ice for 5 min to allow the tissue fragments to settle, and the supernatants, which contained a single-cell suspension of granuloma cells, were collected. The digestion procedure was repeated three times. The supernatant from the first digest was generally discarded, because it often contained large amounts of tissue debris. The supernatants of the second through fourth digests were pooled, and the granuloma cells were washed three times in culture medium by centrifugation at 200 x g for 10 min at 4°C. Remaining tissue debris in the cell suspension was removed by centrifugation over Ficoll-Paque PLUS (Pharmacia, Uppsala, Sweden). The granuloma M were collected, washed three times in culture medium, and counted. These footpad granuloma M are viable and functional cells (e.g., adherent, phagocytic, nonspecific esterase positive, and support the growth of Toxoplasma gondii) (11). Furthermore, microscopic analyses of the cell suspensions and cell-free supernatants verified that 99% of the M. leprae are located intracellularly.
Effector/target M coculture experiments
Normal or ACT (500 U/ml recombinant murine IFN- (R&D Systems) and 5–10 ng/ml LPS (Sigma-Aldrich) for 24 h at 37°C) peritoneal or bone marrow-derived M were overlaid with 5 x 104 target M (M. leprae-infected footpad granuloma M). All effector/target M cocultures were incubated at 33°C, which is the optimum temperature for maintaining M. leprae viability in vitro both axenically (25) and in cultured mammalian cells 4 (27, 28). In experiments analyzing the requirement for cell-to-cell contact, target M were placed in Transwell inserts (4-µm pore size; Corning) and placed over the effector M cultured on coverslips in the wells of the 24-well plate.
Reagents
L-N6-(1-Iminoethyl)lysine hydrochloride (L-NIL) and NG-monomethyl-L-arginine (L-NMA) were obtained from Sigma-Aldrich and ChemBiochem (Salt Lake City, UT) respectively. The concentration of nitrite in the culture supernatants was determined using the Griess reagent (29). Abs to TNF- (clone MP6-XT3) and IFN- (clone XMG1.2) and an isotype control Ab (clone R3-34) were obtained from BD PharMingen (San Diego, CA). ELISA kits for the detection of TNF-, IFN-, and IL-2 in the culture supernatants were obtained from R&D Systems.
Determination of M. leprae viability
The viability of M. leprae recovered from the M cultures was ascertained by radiorespirometry, which measures the oxidation of [14C]palmitic acid to 14CO2, as described previously (10). Briefly, the adherent M were lysed in 200 µl 0.25% SDS (Sigma-Aldrich) to release intracellular M. leprae. After addition of an equal volume of RPMI 1640 plus 20% FBS, 300 µl of lysate was transferred to a 6-ml screw-cap vial containing 4 ml of commercially prepared BACTEC7H12B medium (BD Biosciences, Mountain View, CA) plus 5 µg/ml ampicillin (Sigma-Aldrich) and 2.5 µg/ml amphotericin B (Sigma-Aldrich). The vials, with caps loosened, were placed in wide-mouthed scintillation vials containing a hollow cylinder of Whatman no. 1 filter paper that had been dipped previously in concentrated liquid scintillation counting solution (Liquiflor PPO-POPOP [2,5-diphenyloxazole-1,4-bis(5-phenyloxazolyl)benzene] toluene concentrate; Biotechnology System NEN Research Products, Boston, MA), dried, and charged with 100 µl of 2 N NaOH. The caps of the scintillation vials were firmly tightened, and the assembly was incubated at 33°C. After 7 days, cumulative amounts of oxidized palmitic acid released as 14CO2 by metabolically active M. leprae were measured using a Beckman LS6000ic scintillation counter (Beckman Coulter, Fullerton, CA). Data obtained in this radiorespirometry assay show strong correlation with M. leprae viability as determined in the mouse footpad growth assay (25) and in viability staining assays. 5
Histopathology
The feet from M. leprae-infected nu/nu mice were fixed in 10% buffered formalin, decalcified, and embedded in paraffin. Four-micrometer sections were prepared of cross-sections at the distal, mid, and proximal areas of the metatarsals of the infected foot. The sections were stained with H&E and Fite’s acid-fast stain.
Flow cytometry
The footpad granuloma cells were incubated with rat anti-mouse CD16/CD32 (BD PharMingen) for 10 min at 4°C to block FcRs and then stained for cell surface markers for 30 min at 4°C using Abs against CD3 (anti-CD3 molecular complex
FITC; clone 17A2), CD45R/B220 (anti-CD45R/B220PerCP; clone RA3-6B2), NK (anti-NKPE; clone 2B4), and Mac-1 (anti-CD11bFITC (Mac-1); clone M1/70). Each Ab was titered for optimal cell surface staining, and the appropriate isotype control Abs were used at the same concentrations. All Abs were purchased from BD PharMingen. After staining, the cells were washed, fixed overnight in 1% paraformaldehyde, and analyzed using a FACScan (BD Biosciences). Cells were gated on the lymphocyte and M populations based on size (forward-angle light scatter) and density (side-angle light scatter).
Light microscopy
A total of 1 x 105 granuloma M was centrifuged onto a slide using a Cytospin 2 (Thermo Shandon, Pittsburgh, PA) at 140 x g for 5 min at room temperature. The cells were processed for acid-fast staining by fixation with 10% formaldehyde in ethanol and staining using the Difco BBL TB Stain kit (BD Biosciences) or processed for differential staining with Diff-Quik reagents (American Scientific Products, McGraw Park, IL). Images were captured on a Zeiss Axioplan microscope (Zeiss, Oberkochen, Germany) using a Spot RT camera and software (Diagnostic Instruments, Sterling Heights, MI).
Fluorescence microscopy
Fluorescent effector M.
Fluorescent peritoneal M were obtained from either EGFP mice or from B6 mice labeled using the PKH67 Green Fluorescent Cell Linker mini-kit (Sigma-Aldrich). Briefly, 5 x 107 peritoneal cells were stained and washed according to the manufacturer’s instructions. Peritoneal cells were plated at 2 x 106 cells per culture in 35-mm glass-bottom petri dishes (Electron Microscopy Sciences, Fort Washington, PA), and nonadherent cells were removed by washing after overnight incubation at 37°C.
Fluorescent M. leprae.
Freshly harvested viable M. leprae were labeled using the PKH26 Red Fluorescent Cell Linker mini-kit (Sigma-Aldrich). Briefly, 4 x 109 bacteria were stained and washed according to the manufacturer’s instructions and resuspended in 0.2 ml of PBS. Staining with this dye has no detrimental effect on M. leprae viability as measured by radiorespirometry and growth in the mouse footpad. 5 Fluorescent PKH26-labeled M. leprae was inoculated into the granulomatous footpads of nu/nu mice 11 mo after initial infection with unlabeled viable M. leprae. The granuloma M were harvested 1 wk later and used as target cells.
Microscopy.
Target M infected with PKH26-labeled M. leprae were cocultured with EGFP effector M for 5 days. The cocultures were washed three times in PBS and mounted in 10% glycerol in PBS containing 2 µg/ml each of minocycline (Sigma-Aldrich), ofloxacin (Sigma-Aldrich), and rifampin (Sigma-Aldrich). Confocal images were obtained using a Nikon Eclipse TE-2000E confocal microscope (Nikon Instruments, Lewisville, TX). Fluorescent images were also captured using live targets containing PKH26-labeled M. leprae cocultured with EGFP- or PKH67-labeled effector M for 5 days. These cultures were rinsed three times and placed in PBS before viewing with a Zeiss Axiovert 405M microscope (Zeiss) with a Spot RT camera and software. Images were superimposed using Adobe Photoshop (Adobe Systems, San Jose, CA).
Statistical analyses
All statistics were performed using unpaired t tests in GraphPad InStat Software, version 3.00 (GraphPad Software, San Diego, CA). Data were considered significant at p 
0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inoculation of M. leprae into the footpads of athymic nu/nu mice resulted in an enlarged footpad (Fig. 1A1) comprising a lepromatous-type lesion (A2). Using a series of gentle digestions with collagenase and DNase, we obtained an average of 14.89 x 106 granuloma M (n = 5; range, 7.55 x 106 to 27.2 x 106, depending on the size of the footpad) (Fig. 1A3). These M were engorged with M. leprae and contained an average of 120 ± 17.7 AFB per cell (Fig. 1A4). Flow cytometric analyses of these footpad cell preparations demonstrated that they were composed primarily of Mac-1+ cells (Fig. 1B4). CD3+ cells (Fig. 1B1), B220+ cells (B2), and NK cells (B3) constituted 0.37, 0.21, and 0.05% of the population, respectively.
|
To determine the concentration of M for optimal effector and target cell interaction, M. leprae-infected footpad granuloma target M were cultured alone or cocultured with normal or ACT M at various E:T ratios. As shown in Fig. 2A, an E:T ratio of 1.25:1 or 2.5:1 was not sufficient for killing of M. leprae by ACT effector M. However, an E:T ratio of 5:1 or greater yielded strong inhibitory activity by ACT M while M. leprae metabolism in target M cocultured with an equal number of normal effector M remained high. In fact, coculture of target M with normal effector M at a ratio of at least 2.5:1 consistently yielded significantly higher metabolic activity in the recovered M. leprae.
|
To determine the optimal incubation period for effector and target M coculture, M. leprae-infected footpad granuloma target M were cultured alone or cocultured with normal or ACT M at an E:T ratio of 20:1 for various lengths of time. As shown in Fig. 2B, the metabolic activity of M. leprae harvested from the target-only cultures remained constant over the 5-day cultivation period. Coculture of target M with ACT effector M for up to 48 h resulted in little change in M. leprae metabolic activity. However, 72 h of coculture yielded a significant reduction in M. leprae metabolism that was further reduced after 5 days of coculture. Again, coculture of target M with normal M enhanced M. leprae metabolic activity.
Inhibition of intracellular M. leprae by ACT M is independent of reactive oxygen intermediates (ROI) but dependent on RNI
To determine the importance of RNI and ROI in effector and target cell interaction, M. leprae-infected footpad granuloma target M were cultured alone or cocultured with normal or ACT effector M from B6 or phox91–/– KO mice in the presence or absence of L-NIL, an inhibitor of the L-arginine-dependent production of RNI, or normal or ACT effector M from NOS2–/– mice. The viability of M. leprae as well as the levels of nitrite in the culture supernatants was assessed. As shown in Table I, M. leprae recovered from target M cultured alone exhibited high metabolic activity, and very low levels of nitrite were detected in the culture supernatants. Likewise, target M cocultured with normal B6 effector M contained low nitrite concentrations; however, M. leprae metabolic activity was significantly higher in these cocultures. In contrast, high levels of nitrite were found in the supernatants from target M cocultured with ACT B6 M, which inversely correlated with the low M. leprae metabolic activity in these cocultures. In the presence of L-NIL, nitrite production in the cultures of target M cocultured with ACT effector M was inhibited, and the metabolic activity of M. leprae remained high. Incubation of cocultures of target M and ACT effector M in the presence of 500 µM L-NMA, another inhibitor of RNI production, also prevented killing of M. leprae and nitrite production (data not shown). Results comparable with those obtained with B6-derived effector M were obtained when M. leprae-infected footpad granuloma target M were cocultured with effector M obtained from phox91–/– mice (Table I). The importance of reactive nitrogen products was further substantiated using M. leprae-infected target M cocultured with NOS2–/– effector M. As shown in Table I, ACT NOS2–/– effector M were incapable of generating high levels of nitrites and could not kill M. leprae in target M. In fact, both normal and ACT effector M from this strain of mice enhanced M. leprae metabolic activity.
|
nor IFN- are required during coculture
To determine whether IFN- or TNF- were required for effector M to modify the metabolic activity of target M-derived M. leprae, the cocultures were incubated in the presence of Abs to these cytokines. As shown in Fig. 3A, M. leprae-infected nu/nu footpad granuloma target M produced high levels of TNF-, which was significantly reduced upon cocultivation with both normal and ACT effector M. In the presence of anti-IFN-, the levels of TNF- generated by target M was reduced 40%, but was still higher than that produced in the cocultures. Negligible amounts of TNF- were seen in all cultures in the presence of anti-TNF-. Conversely, low but measurable amounts of IFN- were seen in the cocultures, primarily of target M and ACT effector M (Fig. 3B), and the levels of this cytokine were reduced in the presence of anti-IFN-. However, the presence or absence of these Abs had no effect on the viability of M. leprae or on the amount of nitrite generated compared with control Ab. In the presence of each Ab, cocultivation of M. leprae-infected target M with normal effector M augmented M. leprae metabolic activity, whereas cocultivation with ACT effector M decreased bacterial metabolism (Fig. 3C); again, M. leprae viability was inversely correlated to nitrite levels (D).
|
To determine whether cell-to-cell contact was required for optimum effector and target cell interactions, M. leprae-infected footpad granuloma target M were cultured in Transwell inserts that were overlaid onto effector M monolayers. As shown in Fig. 4, target M cultured on Transwell inserts yielded M. leprae with high viability. When target M were cultured over normal effector M, M. leprae metabolism remained high; however, there was no augmentation of metabolic activity like when the target and effector M were in close contact (Figs. 2 and 3C, Table I). M. leprae-infected target M cultured over ACT effector M also yielded viable M. leprae. This inability of the ACT effector M to kill the target cell-derived M. leprae occurred even though elevated levels of nitrites were generated by these effector cells.
|
with effector M
Granuloma M containing fluorescent PKH26-labeled M. leprae were isolated from the footpads and placed in culture as targets with either green PKH67-labeled normal (Fig. 5A) or ACT (B) effector M or EGFP-derived normal (C) or ACT (D) effector M and incubated for 5 days at 33°C. Imaging by conventional fluorescence microscopy with computer-assisted image overlay (Fig. 5, A and B) demonstrated that the target M-derived bacilli were acquired by both the normal and ACT effector M. These observations were confirmed using confocal microscopy. Images of a single confocal plane (Fig. 5, C and D) verify that the bacilli are inside the effector M. As shown above, under these conditions, the metabolic activity of M. leprae was greatly inhibited in cocultures of infected target M with ACT effector M and bolstered when infected target M were cocultured with normal effector M.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
heavily infected with M. leprae and uninfected M in vitro. Thus, the relevance of these studies focuses on the lepromatous end of the immunohistological spectrum where the disease evolves slowly and is characterized by the gradual accumulation of enormous numbers of bacilli in a local environment that is relatively free of T cells in a host incapable of mounting a specific CMI to the leprosy bacillus. By developing an in vitro model to follow the fate of M. leprae in an LL lesion, we sought, first, to study the host cell (M) dynamics required to maintain and advance the course of clinical disease and, second, to model immunotherapeutic intervention and determine whether ACT effector M alone could interact with the infected target M and have an adverse effect on the viability of the M. leprae residing therein.
Previously, we have shown that granuloma M can be isolated from the M. leprae-infected footpads of the athymic nu/nu mouse (11, 12). We have now optimized this procedure such that we can routinely obtain large numbers of viable, mature, M. leprae-infected M in a suspension that is virtually free of extracellular bacilli. Use of these granuloma M as our target cells has three major advantages. First, these cells differentiated, matured, and became infected in vivo, and are collected directly from the microenvironment of a lepromatous-type lesion. These characteristics make them a highly relevant M population for immunological studies in LL. Second, because the granuloma M are from nu/nu mice and the effector cells are prepared as pure adherent M, we have a system allowing investigation of M interaction without contamination with potentially CTLs (30, 31). Third, to coculture target M with effector M, it is necessary to have at least one of these populations prepared as a suspension culture. Our gentle digestion procedure eliminates the widely used but undesirable and potentially membrane-damaging isolation step of scraping adherent target or effector M from culture dishes to obtain M in suspension. This model is thus appropriate for this initial study of the dynamics of the LL granuloma microenvironment.
Although these granuloma M are heavily burdened with bacilli, they are viable and functional cells that display many characteristics of normal uninfected M in that they are adherent, phagocytic, and express FcRs and the Mac1 Ag; are nonspecific esterase positive; support the growth of T. gondii (11); and, as shown in this manuscript, maintain the viability of M. leprae. However, these M do have a key defect. Unlike peritoneal or bone marrow-derived M, which will phagocytize and kill reasonable numbers of M. leprae if ACT with IFN- (29, 32), the M. leprae-engorged granuloma M are refractory to activation by IFN- and thus manifest aberrant effector functions, including impaired microbicidal and tumoricidal capacity, decreased oxidative metabolic state, and lowered MHC class II Ag expression (11, 12). This defect appears to be the consequence of long-term infection with a high bacillary load because resident peritoneal M infected with large numbers of M. leprae in vitro for several days also become unresponsive to IFN- (13). The mechanism of this down-regulatory effect by M. leprae is not clear; however, subversion of IFN- responsiveness is a survival technique that has been used by other intracellular pathogens (33, 34, 35). The inability of IFN- ACT human M to kill M. tuberculosis has been attributed to this pathogen’s ability to disrupt IFN- signal transduction (35), impair CD64 transcription and surface expression (36), and induce IL-6 production (37). M. leprae-infected M produce copious amounts of PGE2, the presence of which correlates both with M in vitro unresponsiveness to IFN- activation (12) and successful T cell adoptive transfer into M. leprae-infected nu/nu mice (38).
The in vitro studies presented here support previous in vivo observations. Kinetic experiments in nu/nu mice following the traffic of labeled promonocytes into the infected footpad determined that 15–20% of the M. leprae-infected granuloma M were <5 days old (39). Furthermore, IFN- treatment of these mice significantly enhanced the infiltration of M into the footpad granuloma because 25–35% of the M. leprae-burdened M were now newly arrived cells. Thus, the footpad granuloma is a highly dynamic entity containing numerous cells with continuous turnover and replacement of infected M with fresh M, and this dynamic nature can be manipulated experimentally with cytokine treatment. Because the M. leprae-engorged M cannot become ACT by IFN-, these data imply that any killing, breakdown, or clearance of M. leprae from highly bacilliferous tissues resulting from immunotherapeutic or chemotherapeutic intervention would likely be accomplished by the newly arrived, competent M ACT before or shortly after their traffic into the lesion.
Cytokine immunotherapy in leprosy patients further substantiates a dynamic nature for the M. leprae-induced granuloma. Enhancement of CMI and upgrading of clinical and histopathological classification has been attempted in borderline lepromatous and LL patients via cytokine immunotherapy. Upon injection of IFN- or IL-2 into leprosy lesions, a marked cellular infiltration occurred (23, 24, 40), and the bacillary index, a measure of M. leprae burden, was reported to markedly decrease (41). Thus, treatment of lesions with Th1 cytokines apparently evoked a CMI response, but the effect was transient; the specific unresponsiveness to M. leprae Ags characteristic of multibacillary disease was not reversed. It was proposed that the cytokine treatments, the IL-2 treatments in particular, may have induced a population of lymphokine-activated killer (LAK) cells that were responsible for the destruction of the M. leprae-infected M (42). Studies in experimental leprosy have also demonstrated that NK and LAK cells can lyse M. leprae-infected M (43, 44). However, it is unlikely that granuloma-derived NK cells played a role in our in vitro system. Flow cytometric analyses revealed that NK cells constituted only a very minor population of our granuloma cell preparations (NK:M, <0.001:1), a ratio well below that needed for effective NK or LAK cell lysis of mycobacteria-infected target cells (44, 45, 46). Furthermore, no IL-2, the cytokine that promotes differentiation of NK cells into LAK cells, was detected in the supernatants of our cocultures (data not shown).
We have previously shown that the ability of the ACT M to inhibit M. leprae metabolism is highly dependent on the generation of RNI 5 (29, 32). The current studies demonstrate that the inhibitory effects exerted by the ACT effector M were also dependent on the production of RNI. In contrast, killing of the target cell-derived bacilli did not require a phagocyte oxidase-dependent respiratory burst. Peroxinitrite, formed by the interaction of NO and superoxide (47), is a potential product whereby ACT M may exert their antimicrobial effects. However, the inability to generate superoxide by either the ACT phox91–/– effector M (48, 49) or the M. leprae-infected footpad granuloma target M (11, 12) would argue against peroxinitrite as the active toxic molecule in our system.
Although NO is a freely diffusible product of ACT M, its reactivity depends largely on its concentration and the close proximity of the target cells (50). Our studies show that, in addition to RNI, intimate cellular contact between the M. leprae-infected target M and the ACT effector M was required. If the target and effector M were physically separated using Transwell inserts, M. leprae retained high metabolic activity even in the presence of ACT M generating high levels of RNI. Interestingly, cell-to-cell contact was also requisite for the enhancement of M. leprae metabolism by normal effector M.
The importance of TNF- in cellular recruitment to the site of infection, and subsequent granuloma formation and maintenance in response to mycobacterial infection has been demonstrated in numerous investigations (51, 52, 53, 54, 55), and TNF- is expressed across the leprosy spectrum, both in human (56) and experimental (Ref. 27 ; L. B. Adams, N. A. Ray, D. M. Scollard, and J. L. Krahenbuhl, manuscript in preparation) leprosy. In the present study, M. leprae-infected granuloma M synthesized large amounts of TNF- upon ex vivo culture. TNF- levels were reduced, however, upon coculture with effector M. A likely explanation is that the TNF- generated by the target M was bound by receptors expressed by both normal and ACT effector M. Interestingly, addition of anti-TNF- Ab, although blocking TNF- reactivity, did not inhibit the enhancing or detrimental effects of normal or ACT effector M, respectively, on target M-derived M. leprae, nor did it affect the levels of nitrite generated.
Although there were no T cells in our system, production of IFN- was evaluated, because several groups have reported that murine M can produce IFN- under certain conditions. Treatment of M with LPS (57) or IL-12 and IL-18 (58) induces both IFN--specific mRNA expression and protein production. Stimulation with IFN- itself also resulted in the generation of IFN- by M, presumably by an autocrine mechanism (59). Furthermore, Wang et al. (60) demonstrated an IL-12-dependent production of IFN- by alveolar M from Mycobacterium bovis bacillus Calmette-Guérin-infected mice. In our system, M. leprae-infected granuloma M did not themselves generate IFN- in vitro. However, low levels of IFN- were found in the cocultures, especially in the cocultures of target granuloma M with ACT effector M. Even though granuloma footpad target M are refractory to exogenously added IFN-, the possibility that they may become ACT under our coculture conditions was addressed. This IFN- did not contribute to the inhibition of M. leprae metabolic activity, because addition of anti-IFN- Ab, although blocking IFN- reactivity, did not reverse killing of target cell-derived bacilli.
Especially noteworthy is the acquisition of M. leprae by the normal effector M and the subsequent augmentation of M. leprae metabolic activity. M. leprae multiplies essentially unchecked in LL, and it has been postulated that M turnover in the lepromatous lesions results from uptake of bacilli that have been released from infected M that have lysed due to the overwhelming bacterial load in the cells. However, results presented in this study suggest that normal effector M may play a more active role in M turnover. M. leprae have an extremely slow growth rate in vivo and do not multiply in vitro 4 (61). This was confirmed by the relatively constant metabolic rate of M. leprae harvested over the 5-day cocultivation period from our target cell-only cultures. Furthermore, this constant metabolic activity indicates that there was no spontaneous lysis of the M. leprae-infected target M, because this would have resulted in a decrease in recovered bacilli and a drop in metabolic activity over time. Interestingly, even ACT M, in the presence of RNI inhibitors or if NOS2–/– derived, could augment M. leprae viability. Whether or not a succession of challenges with normal M could sustain M. leprae viability in vitro for a prolonged period of time is an intriguing possibility.
The current understanding of the role of CMI in paucibacillary leprosy is drawn from both human and animal studies and suggests that cytotoxic CD4+ and/or CD8+ T lymphocytes or NK and LAK cells destroy the infected and incapacitated host cell (e.g., M or Schwann cell) and release the intracellular bacilli (14, 15, 16, 44). However, in the aftermath of the lysis of the M. leprae-infected cell, there is little direct evidence for the fate of the released bacilli or for a continued role for the M. Pathogens, including mycobacteria, may be killed directly by products of CTLs. Granulysin, a protein in the granules of T cells and NK cells from humans (there is no homolog for granulysin in mice), is lytic against a variety of tumor cells (62) and has been shown to directly kill extracellular M. tuberculosis and, after delivery by perforin, intracellular M. tuberculosis as well (63, 64). Evidence has been presented for granulysin-expressing T cells in skin lesions of patients with TT disease (65), although no direct evidence for killing M. leprae was presented. But we emphasize that the present report is concerned with a model for polar LL, where M. leprae-specific T cells are lacking entirely and target M contain hundreds of bacilli. Although cytotoxic T cell participation is not explored in the present study, our findings support an important M-mediated event in the microenvironment of the granuloma. Upon destruction of the infected M by CTLs, the bacilli are released into the extracellular space where they can be rephagocytized by new, normal M that provide a fresh habitat for them, or competent, ACT M that are capable of killing them (66). This fundamental process could be a key immunoregulatory phenomenon that contributes to fluctuations in the unstable borderline area of the leprosy spectrum.
We do not yet know the mechanism whereby, in our T cell-free in vitro system, effector M acquire the bacilli from the target M or whether normal M acquire the bacilli by a mechanism different from ACT M. One possible scenario could be the lysis of the target M by the effector M and phagocytosis of the released bacilli. ACT M have been shown to be cytotoxic toward tumor cells via an RNI-dependent mechanism (21, 22). An alternate possibility could be the engulfment of infected target M by the effector M. Munn and Cheung (67) have demonstrated both Ab-dependent and Ab-independent phagocytosis of tumor cells by M. A third mechanism of action could be the induction of apoptosis or necrosis in the target M and the phagocytosis of the infected apoptotic/necrotic cells (68, 69). Perskvist et al. (70) have demonstrated the uptake of M. tuberculosis-infected apoptotic neutrophils by M, and Fratazzi et al. (71) reported the adherence of uninfected M to Mycobacterium avium-infected apoptotic M and the subsequent inhibition of mycobacterial growth. Goldmann et al. (72) have shown that dendritic cells phagocytize bacillus Calmette-Guérin-infected necrotic M and present Ag to specific T cells. These possibilities are currently under investigation. Nevertheless, our data show that the state of the M infiltrating the granuloma may markedly affect the viability of M. leprae residing in M in the leprosy lesion.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Linda B. Adams, Immunology Research Department, National Hansen’s Disease Programs, Laboratory Research Branch, Louisiana State University School of Veterinary Medicine, Skip Bertman Drive, Baton Rouge, LA 70803. E-mail address: ladams1{at}lsu.edu
3 Abbreviations used in this paper: LL, lepromatous leprosy; CMI, cell-mediated immunity; AFB, acid-fast bacillus; TT, tuberculoid leprosy; M, macrophage; ACT, activated; RNI, reactive nitrogen intermediate; ROI, reactive oxygen intermediate; EGFP, enhanced green fluorescent protein; L-NIL, L-N6-(1-iminoethyl)lysine hydrochloride; L-NMA, NG-monomethyl-L-arginine; LAK, lymphokine-activated killer.
4 Y. Fukutomi, M. Matsuoka, F. Minagawa, S. Toratani, G. McCormick, and J. Krahenbuhl. Subversion of macrophage antimicrobial function bolsters intracellular survival of Mycobacterium leprae. Submitted for publication.
5 R. Lahiri, B. Randhawa, and J. L. Krahenbuhl. Development of a viability staining method for Mycobacterium leprae derived from the athymic (nu/nu) mouse footpad. Submitted for publication.
Received for publication December 22, 2003. Accepted for publication April 7, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T-cell receptors accumulate in human infectious disease lesions. Nature 339:544.[Medline]
-interferon. Infect. Immun. 55:446.-interferon in macrophages infected with Mycobacterium leprae. Infect. Immun. 56:1912.-interferon in lepromatous leprosy patients. Am. J. Pathol. 128:345.[Abstract]
. J. Exp. Med. 172:1509. gene-disrupted mice as a model to explore the immunopathologic spectrum of leprosy. J. Infect. Dis. 185:S1.
-interferon-induced tyrosine phosphorylation in mononuclear phagocytes infected with Lieshmania donovani: selective inhibition of signaling through Janus kinases and Stat1. Infect. Immun. 63:4495.[Abstract]
transcriptional responses without inhibiting activation of STAT1. J. Immunol. 163:3898. without inhibiting STAT1 function. J. Immunol. 171:2042.. J. Immunol. 171:4750.-Interferon in experimental leprosy. Diagn. Microbiol. Infect. Dis. 13:405.[Medline]
injections on the disposal of Mycobacterium leprae. Proc. Natl. Acad. Sci. USA 86:8073. in Mycobacterium-induced granuloma formation in TNF-deficient mice. Lab. Invest. 79:379.[Medline]
in macrophages by lipopolysaccharide. Int. Immunol. 5:1383. upon combined stimulation with interleukin (IL)-12 and IL-18: a novel pathway of autocrine macrophage activation. J. Exp. Med. 187:2103. upregulates its own gene expression in mouse peritoneal macrophages. J. Exp. Med. 179:1731.This article has been cited by other articles:
|
|
 |
|
  T. Mukai, Y. Maeda, T. Tamura, M. Matsuoka, Y. Tsukamoto, and M. Makino Induction of Cross-Priming of Naive CD8+ T Lymphocytes by Recombinant Bacillus Calmette-Guerin That Secretes Heat Shock Protein 70-Major Membrane Protein-II Fusion Protein J. Immunol., November 15, 2009; 183(10): 6561 - 6568. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Hagge, B. M. Saunders, G. J. Ebenezer, N. A. Ray, V. T. Marks, W. J. Britton, J. L. Krahenbuhl, and L. B. Adams Lymphotoxin-{alpha} and TNF Have Essential but Independent Roles in the Evolution of the Granulomatous Response in Experimental Leprosy Am. J. Pathol., April 1, 2009; 174(4): 1379 - 1389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Williams, T. L. Pittman, M. Deshotel, S. Oby-Robinson, I. Smith, and R. Husson Molecular Basis of the Defective Heat Stress Response in Mycobacterium leprae J. Bacteriol., December 15, 2007; 189(24): 8818 - 8827. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Makino, Y. Maeda, and K. Inagaki Immunostimulatory Activity of Recombinant Mycobacterium bovis BCG That Secretes Major Membrane Protein II of Mycobacterium leprae. Infect. Immun., November 1, 2006; 74(11): 6264 - 6271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Scollard, L. B. Adams, T. P. Gillis, J. L. Krahenbuhl, R. W. Truman, and D. L. Williams The Continuing Challenges of Leprosy Clin. Microbiol. Rev., April 1, 2006; 19(2): 338 - 381. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or cocultured with various numbers of normal (
) or ACT (
) effector M
