Aminopeptidase N (CD13) is an ectoenzyme located in the outer membrane of a variety of cells. Proteomic profiling indicates an increased expression of CD13 in phagocytes during Mycobacterium tuberculosis infection. The purpose of this study was to investigate the role of CD13 on the internalization and intracellular survival of M. tuberculosis in monocytes.
Magnetic nanoparticles and confocal microscopy were used to observe interactions between CD13 and M. tuberculosis. Mycobacterial entry and intracellular survival in monocytes were assessed with and without anti-CD13 antibody (WM15 and WM47) using flow cytometry and colony formation assay.
By using magnetic nanoparticles and confocal microscopy, M. tuberculosis was found to be capable of binding to either soluble CD13 or membranous CD13 on monocytes. Flow cytometry showed that pretreatment of monocytes with WM15 or WM47 reduced the number of intracellular M. tuberculosis. Collectively, the data suggest that CD13 is a binding and entry receptor for M. tuberculosis on monocytes. Treatment of infected monocytes showed a greater effect of WM47 than WM15 in reducing the intracellular colonization of M. tuberculosis, suggesting that specific epitopes of CD13 may play an important role modulating intracellular M. tuberculosis survival.
CD13 acts as a receptor for M. tuberculosis on human monocytes. The molecule facilitates internalization, and interaction of CD13 with an anti-CD13 antibody reduces intracellular M. tuberculosis survival.
Mycobacterium tuberculosis is one of the most successful pathogens estimated to have infected nearly one-third of the human population and killing approximately 1.7 million people each year. Much of its success is due to essential virulence factors that allow it to survive within phagocytes rather than to be eliminated by these scavenger cells. Mycobacteria bind to macrophages in cholesterol and lipid-rich domains of the host cell plasma membrane called lipid rafts, which are associated with various signalling mechanisms. Once within the cell, the organism can degrade cholesterol as energy source to maintain a chronic infection in the host. Additionally, mycobacterial interference with lipid-mediated signalling arrests phagosome maturation, thus protecting the bacterium against delivery to the lysosome.
Aminopeptidase N (CD13) is a multifunctional protein expressed in many tissues and has been found to be partially localized in lipid rafts. It influences plasma membrane protein organization and cholesterol uptake. It can exist both as a membrane-bound protein and as an active soluble protein secreted by certain cells or released by cleavage of the plasma membrane. CD13 has been shown to be a cell-surface receptor for certain viruses, seemingly required for endocytosis enabling internalization of the virus into the cell.[9, 10] By a modified isotope-coded affinity tag technology, we previously found a significantly elevated expression of CD13 in human phagocytes infected with M. tuberculosis (Lu & Tsai, unpubl. data, 2007). We therefore speculated that CD13 might have a yet undefined role in mycobacterial infection. This study was designed to assess the interaction between CD13 and M. tuberculosis, investigating both entry of the organism into monocytes and its subsequent intracellular survival.
This study was conducted according to the principles expressed in the International Conference on Harmonisation/World Health Organization Good Clinical Practice standards, and written informed consent was obtained for participation in the study, which was approved by the institutional review board of the Mackay Memorial Hospital.
M. tuberculosis preparation
M. tuberculosis strains were obtained from the culture collection of Mycobacteriology Laboratory, Mackay Memorial Hospital, Taipei, Taiwan. The organisms were cultured on Lowenstein-Jensen medium slant at 37°C in a 10% CO2 humidified atmosphere. Detailed documentation of the experimental procedures was described in the Appendix S1 and S2 in the online supporting information.
Cell culture and mycobacterial infection
Peripheral blood mononuclear cells were isolated from the whole blood of healthy adult volunteers by Ficoll-Paque gradient centrifugation. Mononuclear cells were incubated with CD14 microbeads (Miltenyi Biotec, Auburn, CA, USA) and then the CD14-positive cells were separated by means of a magnetic force. These cells were seeded in U-bottom 96-well plates at a density of 2 × 105 cells in a volume of 200 μl of RPMI-1640 medium with 10 % fetal bovine serum (FBS) (Biological Industries, Kibbutz Beit Haemek, Israel) and infected with M. tuberculosis (approximately 5 × 105). Monocytes were washed repeatedly to remove extracellular bacteria and incubated with monoclonal antibodies against CD13 (clone WM15, Biolegend or WM47, Santa Cruz Biotechnology, Santa Cruz, CA, USA) or isotype (mouse IgG1κ, Biolegend, San Diego, CA, USA).
Binding of soluble CD13 to M. tuberculosis was examined by flow cytometry. A suspension of about 1 × 107 bacilli/mL was incubated with recombinant human CD13 (residues 69-967, R&D, Minneapolis, MN, USA) or CD4 (residues 26–226, Abcam, Cambridge, UK) for 30 min at 37°C, washed twice with phosphate buffer saline and centrifuged at 3500 g for 15 min at 4°C. The pellet was resuspended in phosphate buffer saline, after which phycoerythrin-conjugated mouse antibody against isotype (IgG1κ), or CD13 (clone L138) or CD4 (clone RPA-T4) (BD Pharmingen, San Jose, CA, USA) was added for 30 min at room temperature, followed by washing with phosphate buffer saline twice and centrifugation at 3400 g for 15 min at 4°C. The pellet resuspended in phosphate buffer saline was mixed with an equal volume of 4% formalin and incubated for 24 h prior to flow cytometric analysis. For the assay of bacteria entry, intracellular quantities of M. tuberculosis were measured by staining permeabilized cells with fluorescein isothiocyanate-conjugated anti-M. tuberculosis antibody. Monocytes were washed repeatedly to remove extracellular bacteria and incubated with 0.2% trypan blue for 2 min at 4°C to allow efficient quenching of surface fluorophore. The mean fluorescence intensity of stained cells was measured by fluorescence-activated cell sorting Calibur flow cytometry and analysed by CellQuest software (BD Bioscience, San Jose, CA, USA).
Magnetic nanoparticle preparation
Magnetic nanoparticles (MNP) composed of Fe3O4 were pre-labelled with nitrilotriacetic acid and conjugated with recombinant histidine-tagged CD13 protein. MNP composed of Fe3O4 coated with nitrilotriacetic acid derivative was kindly donated by Professor Yu-Chie Chen, National Chiao Tung University, Hsinchu, Taiwan. CD13-MNP was prepared by pre-labelling the surface of MNP with Ni(II), which was then conjugated to recombinant histidine-tagged CD13 protein through the binding to Ni(II). M. tuberculosis (108 bacilli/mL, 200 μL) were mixed with either unbound nanoparticles or CD13-MNP (0.25 μg/mL, 100 μL) and then allowed to precipitate for 10 min. By applying an external magnetic field, nanoparticles and the attached mycobacteria were attracted to the tube wall. The amount of unbound M. tuberculosis was estimated by measuring the absorbance at 600 nm with a spectrophotometer. The nanoparticles were smeared onto slides and acid-fast stain was performed. The slides were observed under the microscope.
Monocytes were cultured on 18-mm diameter cover glass placed in 12-well culture plate and infected with M. tuberculosis labelled with Auramine-Rhodamine T. After 30 min, unbound bacteria were washed away with phosphate buffer saline and monocytes were fixed in 4% formalin. CD13 were stained with Cy-Chrome 5-conjugated anti-CD13 antibody, and nuclei were stained with 4′-6-diamidino-2-phenylindole. Samples were analysed by Leica true confocal scanner SP5 confocal laser scanning microscopy (Leica Microsystems, Wetzlar, Germany).
Colony forming units
Cells incubated with M. tuberculosis were lysed by adding 100 μL 0.1% sodium dodecyl sulfate, vortexing and incubated for 10 min. Additional 900 μL H2O was added to the solution and centrifuged at 3400 g for 15 min at 4°C. The lysates were plated on 7H11 agar and incubated for 3 weeks. The resulting growth of M. tuberculosis was reported as mean colony forming units (CFU) per 10 000 cells.
Paired t-test was used for analysis. Data are reported as the mean ± standard error of the mean. Statistical analysis was performed using Prism 3.0 software (GraphPad Software, Inc., San Diego, CA, USA). Two-sided tests were used, and a P-value of <0.05 was considered statistically significant.
Interactions between CD13 and M. tuberculosis
To assess whether M. tuberculosis interacts with CD13, bacilli were incubated with or without soluble recombinant human CD13 protein for 30 min. The binding of soluble CD13 to M. tuberculosis was detected by staining with phycoerythrin-conjugated anti-CD13 antibody and an isotype antibody as negative control. As demonstrated by the flow cytometry, a dose-dependent increase of CD13-positive M. tuberculosis organisms was observed with the binding of CD13 to M. tuberculosis reaching up to 6.53% (Fig. 1a,b). An irrelevant protein CD4 was concurrently utilized to test whether the binding of M. tuberculosis was specific for CD13. No apparent binding was observed between CD4 (0.5 μg) and M. tuberculosis (Fig. 1a, bottom row). CD13-MNP was also prepared to evaluate the binding affinity between CD13 and M. tuberculosis. While applying an external magnetic field to accelerate the process of aggregation, we noticed that CD13-MNP aggregates were lightly coloured in comparison with those of MNP (Fig. 1c). We speculated that abundant M. tuberculosis organisms bound to CD13-MNP shuttered the natural brown colour of MNP. To estimate the amount of unbound M. tuberculosis in the solution, we removed the aggregate pellets by applying an external magnetic field and estimated the density of mycobacteria in the solution by measuring absorbance at 600 nm (Fig. 1d). Prior to the addition of MNP, mycobacterial concentration was 3.7 ± 0.2 × 107/mL. At 10 min after mixing, much less unbound M. tuberculosis (0.4 ± 0.1 × 107/mL) was detected in the CD13-MNP solution than in the MNP solution (2.6 ± 0.2 × 107/mL). Following acid-fast staining, the majority of plain nanoparticle aggregates were found to be small, with only occasional acid-fast organisms seen (Fig. 1e). By contrast, CD13-MNP aggregates occurred in large clusters surrounded by and intermingled with abundant acid-fast organisms. These results indicate that soluble CD13 binds M. tuberculosis.
To find out whether M. tuberculosis interacts with CD13 on the surface of monocytes, M. tuberculosis were incubated with monocytes at different time points and examined by confocal microscopy. At 20 min of incubation with monocytes, M. tuberculosis was mostly found in the extracellular location with the bacteria partly colocalized with CD13 (Fig. 2, top panel). Complete colocalization of M. tuberculosis with the surface CD13 could be found at 30 min of incubation (Fig. 2, middle panel), and the rate of colocalization was 13.1 ± 4.7% (n = 3). At 60 min of incubation, most of M. tuberculosis were found inside the monocytes (Fig. 2, bottom panel). Furthermore, we also tested whether CD13 mediates the binding of mycobateria with human macrophages (Appendix S3 in the online supporting information). Hence, CD13, in both soluble and membrane-bound forms, is capable of binding M. tuberculosis. In other words, CD13 may serve as a receptor to bind M. tuberculosis onto the monocyte and macrophage surface.
CD13 mediates the internalization of M. tuberculosis
To understand whether the expression of CD13 on monocytes is involved in the process of M. tuberculosis internalization, cells were pretreated with two clones of monoclonal antibodies against CD13, WM15 and WM47. Both decrease CD13 expression, but WM15 also inhibits the aminopeptidase activity.[13, 14] To explore whether mycobacterial internalization was mediated through CD13, monocytes pretreated with WM15 and WM47 were then incubated with M. tuberculosis for 24 h. Our results show that the ratio of M. tuberculosis-positive monocytes was significantly reduced by the treatment of 10 μg/mL WM15 (72.1 ± 4.9%, P = 0.0046) or 10 μg/mL WM47 (66.2 ± 5.3%, P = 0.003) as compared with that of isotype control (Fig. 3). As with monocytes, anti-CD13 antibody also reduces the entry of mycobateria into macrophages (Appendix S3 in the online supporting information). Such data support that CD13 is one of the entry receptors on monocytes and macrophages for M. tuberculosis.
The role of CD13 in intracellular survival of M. tuberculosis
To study the effects of CD13 on intracellular mycobacterial growth, monocytes were infected with M. tuberculosis for 1 h followed by 24-h anti-CD13 antibody treatment. After 3-week culture from the monocyte lysates, the CFU counts of M. tuberculosis with WM47 (2635 ± 430, P = 0.0002) as well as WM15 (3248 ± 322, P = 0.049) treatment were significantly less than that with isotype control (3827 ± 483) (Fig. 4). Our data showed that WM47 was superior in suppressing intracellular bacterial growth compared with WM15 (P = 0.0324). Collectively, treatment with anti-CD13 antibodies could reduce not only the entry of M. tuberculosis into monocytes but also the survival numbers of intracellular bacteria. These findings suggest that CD13 may facilitate the internalization of M. tuberculosis into monocytes as well as modulate intracellular survival of the organisms.
The current study employing recombinant soluble CD13, nanoparticle-bound CD13 and membrane-bound CD13 provides evidences that CD13 serves as a receptor on monocytes to bind M. tuberculosis. However, internalization of M. tuberculosis is evidently not dependent on the enzymatic activity CD13, even though this activity is considered an essential biological function of the receptor. Similar results have been reported for in vitro experiments with a human coronavirus and cytomegalovirus infection,[16, 17] showing that CD13-mediated uptake of virus was not dependent on its enzymatic activity.
We also observed that soluble CD13 nanoparticles bound about 6% of the mycobacteria in solution, but treatment of monocytes with anti-CD13 antibodies led to a 30% reduction in the internalization of M. tuberculosis. M. tuberculosis possesses numerous dissimilar ligands on its surface and is therefore likely to engage multiple receptors on phagocytes, including complement receptor 3, mannose receptor, surfactant protein A, class A scavenger receptors on monocytes/macrophages and dendritic cell-specific intercellular adhesion molecule-grabbing non-integrin. Most of these phagocyte receptors are raft-associated and play major or auxiliary roles in the process of binding, phagocyting and transporting M. tuberculosis into the cell.
It has been suggested that the receptors involved in phagocytic entry of the bacilli have a major influence on the pathogens' intracellular survival. For example, complement receptor 3-mediated phagocytosis does not result in the same degree of inflammatory response associated with the invasion of macrophages by M. leprae. Internalization of M. tuberculosis via the mannose receptor generates a negative signal delivered through the mannose receptor and inhibits the phagosome-lysosome fusion. The use of these entry points therefore apparently confers a survival benefit. Nevertheless, Zimmerli et al. found that blocking these two particular receptors with specific antibody did not alter the survival and growth of M. tuberculosis in human macrophages. Our data showed a somewhat different outcome because treatment with two dissimilar anti-CD13 antibodies affected not only the internalization but also intracellular survival of M. tuberculosis in monocytes, particularly the WM47 antibody. This implies that WM47-specific epitope on CD13 may be responsible for reducing intracellular M. tuberculosis survival. It has been shown that cross-linking CD13 with a defined clone of anti-CD13 antibodies induces cell activation, including mitogen-activated protein kinase phosphorylation, calcium-fluxing and homotypic aggregation of monocytes in an epitope-dependent way.[28-30] It seems likely that a series of interactions provoked by the treatment with anti-CD13 may link to intracellular growth inhibition of M. tuberculosis in monocytes.
To examine mechanisms underlying CD13-associated modulation of mycobacterial growth in monocytes, representative bactericidal activities (reactive oxygen species and cytokine production) were examined, but no significant differences were found (Appendix S4 in the online supporting information). One of the strategies used by M. tuberculosis to escape from phagosomes and grow within the cytosol of phagocytes is through modulating the phagosome acidification.[31-33] We found that significantly lower pH value and higher number of active lysosomes (pH < 4.8) were observed in cells treated with WM47 but not with isotype control or WM15 (Appendix S5 in the online supporting information). CD13 has been proposed as a possible receptor of the cholesterol absorption inhibitor ezetimibe, raising the possibility that CD13 participates in the uptake of cholesterol, the important nutrient for intracellular M. tuberculosis persistence. Further studies are needed to clarify these aspects.
In conclusion, this study highlights a putative role for CD13 as a novel monocyte receptor for M. tuberculosis. CD13 is capable of facilitating internalization and intracellular survival of M. tuberculosis in human monocytes. This information adds to our growing understanding about how M. tuberculosis can infect cells and continue to survive within them over a long period of time—and to develop novel anti-tuberculosis therapies.
This project was funded with supports from Mackay Memorial Hospital (Projects # MMH-E-98008 and MMH-E-99008) and Taiwan National Science Council (Grant # NSC 98-2320-B-195-001 and NSC 99-2314-B-195-008). We would like to specially give thanks to Dr Mary Jeanne Buttrey for critical review of the manuscript. We also thank Mr Ya-shiuan Lin for compounding MNP and his assistance in experiments.