- Research article
- Open Access
Mycobacterium tuberculosis infection induces non-apoptotic cell death of human dendritic cells
© Ryan et al; licensee BioMed Central Ltd. 2011
- Received: 11 March 2011
- Accepted: 24 October 2011
- Published: 24 October 2011
Dendritic cells (DCs) connect innate and adaptive immunity, and are necessary for an efficient CD4+ and CD8+ T cell response after infection with Mycobacterium tuberculosis (Mtb). We previously described the macrophage cell death response to Mtb infection. To investigate the effect of Mtb infection on human DC viability, we infected these phagocytes with different strains of Mtb and assessed viability, as well as DNA fragmentation and caspase activity. In parallel studies, we assessed the impact of infection on DC maturation, cytokine production and bacillary survival.
Infection of DCs with live Mtb (H37Ra or H37Rv) led to cell death. This cell death proceeded in a caspase-independent manner, and without nuclear fragmentation. In fact, substrate assays demonstrated that Mtb H37Ra-induced cell death progressed without the activation of the executioner caspases, 3/7. Although the death pathway was triggered after infection, the DCs successfully underwent maturation and produced a host-protective cytokine profile. Finally, dying infected DCs were permissive for Mtb H37Ra growth.
Human DCs undergo cell death after infection with live Mtb, in a manner that does not involve executioner caspases, and results in no mycobactericidal effect. Nonetheless, the DC maturation and cytokine profile observed suggests that the infected cells can still contribute to TB immunity.
- Meso Scale Discovery
- Immunogenic Cell Death
- FITC Mouse
- Macrophage Cell Death
- Victor2 Plate Reader
Tuberculosis is responsible for 1.7 million deaths annually, and Mycobacterium tuberculosis (Mtb) infects up to one third of the world's population [1, 2]. Yet the human host response to Mtb infection in 90% of cases is an immune success story; where infection is followed, not by disease, but by lifelong latent infection .
The key role played by dendritic cells (DCs) in this successful host response has been well studied . After inhalation, Mtb bacilli are phagocytosed by alveolar macrophages and DCs resident in the alveolar space. It falls to the DCs to efficiently travel to local lymph nodes and successfully present antigen to T cells, which generates effective cell-mediated immunity [4, 5]. Dissemination of mycobacteria to the lymph node, which occurs in part via infected DCs, is an important precursor to T cell activation [6, 7]. A deficiency of DCs, monocytes, B and NK cells (DCML deficiency), with an as yet unknown genetic basis, has recently been defined in four subjects. Two of these subjects succumbed to mycobacterial infection: one developed disseminated BCG-osis and the other was diagnosed with spontaneous Mycobacterium kansasii infection . Similarly, mutations in interferon regulatory factor 8 (IRF8), described recently in three subjects, are associated with dendritic cell deficiency resulting in susceptibility to disseminated BCG-osis 
We and others have shown how macrophage cell death follows infection with Mtb [10–13]. This macrophage response has consequences for aspects of innate and cell-mediated immunity [14, 15]. The impact of Mtb infection on DC survival, however, is poorly understood. Given the non-redundant role of DCs in mycobacterial immunity , and their identification as a target for novel therapies and vaccines [4, 16–19], we sought to define the requirements and mechanism of DC cell death after infection with Mtb. By modelling human monocyte-derived DCs in vitro, we infected DCs with Mtb to assess phagocyte survival, and attendant caspase activity, cytokine production and mycobactericidal effect.
Our results show that Mtb infection drives DC maturation and death. As we found in macrophages , the cell death that follows Mtb H37Ra infection is caspase-independent and is not characterised by nuclear fragmentation. In fact, infected DC death proceeds without the activation of caspases. Increased cytokine production followed DC infection with Mtb, but isolated DCs were not able to kill intracellular bacilli. Such data is of value in projecting how manipulation of DCs for new therapeutic strategies can be modelled.
Live M. tuberculosis infection causes dendritic cell death
Maturation was assessed in DCs infected with H37Ra. In controlled experiments, DCs were infected with live or dead Mtb H37Ra or at MOI 1for 24 h. Approximately 60% of cells had phagocytosed mycobacteria at this time point. The cells were washed to remove extracellular mycobacteria and either analysed or incubated for a further 24 or 48 h before analysis. DCs infected with live H37Ra displayed a mature phenotype, up-regulating CD83 and CD86 after 48 h infection with Mtb (Figure 1D). Streptomycin-killed H37Ra did not induce DC maturation.
Having established that reduced DC viability was dependent on infection with live mycobacteria, we then investigated the mechanism of cell death in H37Ra-infected DCs. We previously noted that macrophage cell death after Mtb infection results in DNA fragmentation. By ELISA, we could show that DNA fragmentation was also a feature of the DC response to viable Mtb H37Ra infection peaking at an MOI of 5 (Figure 2D).
Apoptosis results in nuclear condensation, pyknosis and, eventually, fragmentation of the nucleus into apoptotic bodies [20, 21]. To determine whether this occurred during Mtb H37Ra infection, the nuclear morphology of DCs stained with Hoechst was examined by epifluorescent microscopy. The nuclei of infected cells did not undergo pyknosis or fragmentation and were similar in appearance to those of uninfected cells at 72 h after infection, a time at which they had undergone significant cell death. DCs treated with cycloheximide and staurosporine displayed extensive nuclear fragmentation, indicating that the cells are capable of undergoing this process when treated with apoptotic stimuli (Figure 2E).
Dendritic cell death after M. tuberculosis H37Ra infection is caspase-independent and proceeds without the activation of caspase 3 and 7
Our results so far indicated that H37Ra-infected DC death occurred with DNA fragmentation, but without nuclear karyorrhexis and was caspase-independent. Caspase-independent cell death can occur with or without caspase activation, depending on the mechanism of cell death . In order to more closely examine the role of caspases in DC death induced by Mtb H37Ra infection, we analysed the activity of the executioner caspases 3/7 in parallel with cell death at 24 h, 48 h and 72 h post-infection with Mtb (Figure 3C). Staurosporine (24 h treatment at all time points) and cycloheximide (24, 48 and 72 h treatment in parallel with infection) were used as positive controls for caspase activity, inducing increased caspase 3/7 activity at all time points examined (Figure 3C). Caspase activity was measured before and after significant cell death had occurred. Cell death due to Mtb H37Ra was apparent at 72 h post-infection (Figure 3C) and occurred with live Mtb infection only, as in our previous experiments (Figure 2). Caspases 3/7 were not active above levels recorded in uninfected DCs at any time point examined, indicating that these caspases are not activated during DC death after Mtb H37Ra infection.
Secretion of cytokines by Mtb H37Ra-infected dendritic cells
Dendritic cells are permissive for growth of Mycobacterium tuberculosis H37Ra
We investigated the impact of Mtb infection on the viability of human monocyte-derived dendritic cells. We found that DC death followed infection with both the H37Ra and H37Rv strains of Mtb, required viable bacilli, and could be detected at 24 hours co-incubation. The type of cell death was atypical of apoptosis, because it lacked nuclear fragmentation. Cell death due to infection with H37Ra was caspase-independent, although it did proceed with DNA fragmentation. Caspase activation was not detected by substrate assay analysis. Although this type of cell death did not interfere with earlier DC maturation events or cytokine release, it was not associated with any detectable mycobactericidal effect of DCs.
With regard to mycobactericidal effect, DC death differs from H37Ra-infected macrophage cell death, which can kill the invading parasite . In murine DCs the consequences of cell death after infection with Legionella pneumophila link caspase activity and bacterial killing , however we did not see caspase 3 or 7 activity, or association with Mtb killing. Other groups have examined DC mycobactericidal capacity using different models, with differing results [34–36]. Fortsch et al. and Bodnar et al. [34, 35] found that DCs were permissive for growth of intracellular Mtb, while Tailleux et al.  reported constraint of Mtb replication within DCs without the addition of IFN-γ. The proposed difference in findings was suggested to be due to removal of the cytokines GM-CSF and IL-4 from DCs upon infection with Mtb. We maintained the GM-CSF/IL-4 supplementation of our DCs in culture to maintain the DC phenotype, and these factors did not support infected DC viability or ability to limit intracellular bacterial replication. Similar findings were reported in murine Mtb-infected DCs maintained in IL-4, which were unable to control mycobacterial growth in the absence of exogenous IFN-γ . Our experiment models the early stages of Mtb infection in the lung where newly arrived DCs may become infected before being activated by exposure to TH1 cytokines allowing uncontrolled proliferation of mycobacteria. After the initiation of a T cell response and the formation of the granuloma infected DCs are more likely to be exposed to IFNγ and may be better able to control the growth of mycobacteria.
It is perhaps not surprising that DCs failed to kill bacilli by themselves, without T cell help. Other consequences of macrophage cell death after Mtb infection have been described, such as better antigen presentation [14, 37]. DCs play a key role in antigen presentation, which results in activation of T cell populations that can lead to efficient phagocyte killing of the intracellular bacillus, via granulysin-induced phagocyte death, or by cytokine release (e.g. IFN-γ) that supports the mycobactericidal capacity of phagocytes [38–41]. Although outside the scope of this current article, it is possible that dying DCs share some properties of dying macrophages, and contribute to this T cell response.
In the present study we found that both the attenuated H37Ra and virulent H37Rv strains cause death of human DCs. The caspase-independent cell death we report in H37Ra-infected DCs appears to be neither apoptosis nor pyroptosis (both of which require caspase activity) [22, 42]. There are various modes of non-apoptotic cell death, such as pyronecrosis and necroptosis, which can occur without caspase activation. The way in which cells die shapes the response of the immune system; death can be immunogenic, tolerogenic or silent [43, 44]. Therefore, the type of cell death undergone by Mtb-infected DCs is of interest, as it may either support or inhibit cytotoxic and helper T cell responses. Macrophage apoptosis appears to be beneficial for the host response to tuberculosis by having direct bactericidal effects on intracellular mycobacteria and also in the stimulation of protective immunity. The genome of M. tuberculosis contains genes that actively inhibit macrophage apoptosis and enhance its intracellular survival, including nuoG, pknE and secA2 . It is likely that the products of these genes would also inhibit apoptosis of DCs, possibly steering the cells towards the non-apoptotic mode of cell death seen in the present study. Interestingly, foamy macrophages (which are positive for DC markers) in granulomas in the lungs of mice infected with M. tuberculosis have been found to express high levels of TNFR-associated factors (TRAFs) 1-3 which are associated with resistance to apoptosis . Although H37Ra and H37Rv are highly related, being derived from the same parental H37 strain, they differ in important respects at the genetic , transcriptional  and post transcriptional  levels. As a result H37Ra displays several characteristics that are different from H37Rv (e.g. variations in PE/PPE/PE-PGRS proteins , decreased survival inside human macrophages [50, 51], differences in the composition of mannose caps on lipoarabinomannin  and impaired ability to secrete ESAT 6 ) each of which could have an impact on the mode of cell death [53, 54]. Indeed, similar to our previous finding in human macrophages , H37Rv infection killed DCs at a significantly faster rate than H37Ra. Further work will be needed to determine whether infection of DCs with H37Rv causes a similar caspase-independent mode of cell death.
Caspases can have variable effects on the immunogenic potential of dying cells. Exposure or release of damage-associated molecular pattern molecules (DAMPs), such as calreticulin, high-mobility group box 1 (HMGB1) or heat-shock proteins, by dying cells is associated with immunogenic cell death. Calreticulin exposure has been shown to be of particular importance in the induction of immunogenic cell death . Exposure of calreticulin is caspase-dependent; however caspases can also mitigate the pro-inflammatory release of DAMPs from dying cells and cell death that proceeds without the activity of caspases may generate more immune-activating DAMPs [43, 56]. Such an outcome might benefit the host response. These DAMPs could escape from the cell, unimpeded by caspase-neutralisation, and proceed to work in concert with the pro-inflammatory cytokine profile we observed, to generate a better inflammatory response in the lymph node. Yet, cross-priming of T cells is improved by caspase-dependent macrophage apoptosis [14, 57]. Whether DC death that occurs without caspase activation can elicit a CD8+ T cell response remains to be seen.
It is also possible that DC death could interfere with important DC functions such as migration to local lymph nodes for efficient antigen presentation. Others have shown that DC migration to local lymph nodes is impaired in Mtb infection [58, 59], which would delay stimulation of T cell responses. Although DC death could contribute to this phenotype, DC migration to the draining lymph node can take 18 hours in vivo after challenge with Mtb . Although we cannot extrapolate directly from our in vitro experiments to the complex environment that these cell are exposed to in vivo, infected DCs are known to traffic from the lung to lymph nodes . At low MOI, the DC may arrive at the node before undergoing death in an environment where cell death can contribute to antigen cross-presentation. Elimination of the infected DCs could also deprive the host response of an important source of cytokines and antigen presentation; though data from Alaniz et al. suggest that DCs can serve, like macrophages, as a niche cell that promotes intracellular bacterial replication . Mtb-infected DCs produced IL-1β, IL-6, IL-8, IL-10, IL-12p70 and TNF-α as reported previously [62–66] despite the fact that the majority of the cells eventually die. The cytokine profile of Mtb-infected DCs would successfully drive differentiation of TH1 and TH17 responses .
Mtb and the human immune system have co-evolved, so that one third of the global population has been colonised by this pathogen, yet the immune system is adequate at preventing disease 90% of the time [1, 2]. The central cell that regulates this host response is the dendritic cell, and consequently it is increasingly viewed as a target for new therapeutic and vaccine strategies [19, 68]. It is hoped that our description of the DC death response to Mtb infection - as pro-inflammatory, and without the activation of caspases - will inform further research that defines the T cell consequences of this innate response. An improved understanding of these immune processes should help advance translational research to address this global killer.
We have presented evidence that DCs undergo cell death after infection with Mtb in vitro, just as macrophages do. In H37Ra infection this non-apoptotic response does not limit the viability of the infecting bacillus, yet it does not interfere with DC maturation or cytokine production, as previously reported. The lack of caspase activity seen may also contribute to the host response by allowing DAMPS to drive anti-TB immunity, without neutralisation by these important proteases. Further work is needed to determine whether the virulent strain H37Rv induces a similar non-apoptotic form of cell death in human DCs.
M. tuberculosis strains H37Ra and H37Rv were obtained from the American Type Culture Collection (Manassas, VA). Mycobacteria were propagated in Middlebrook 7H9 broth (Difco/Becton Dickinson, Sparks, MD) supplemented with albumin-dextrose-catalase supplement (Becton Dickinson) and 0.05% Tween 80 (Difco). Aliquots were stored at -80°C, thawed and grown to log phase in Middlebrook 7H9 medium before use.
Inactivation of mycobacteria with streptomycin
Log-phase H37Ra were treated with streptomycin sulphate (Sigma, St. Louis, MO; 0.1 mg/ml) for 48 h prior to infection. Streptomycin was thoroughly washed from mycobacteria prior to DC infection.
Obtained through the NIH Biodefense and Emerging Infections Research Resources Repository, NIAID, NIH: Mycobacterium tuberculosis, Strain H37Rv, Gamma-Irradiated Whole Cells, NR-14819.
Peripheral blood mononuclear cells (PBMCs) were isolated from buffy coats of anonymous healthy donors (provided, with permission, from the Irish Blood Transfusion Service). The PPD status of donors was unknown. PBMCs were separated by density centrifugation on Lymphoprep (Axis-Shield, Oslo, Norway), washed and re-suspended in serum-free RPMI 1640 (Gibco, Invitrogen, Carlsbad, CA; for plastic adherence monocyte separation) or in PBS (Sigma) with 2% defined foetal bovine serum (FBS; HyClone, Thermo Fisher Scientific, Waltham, MA) and 1 mM EDTA (Sigma) (for immunomagnetic negative selection). Monocytes were isolated by plastic adherence, or by negative selection using the immunomagnetic negative selection EasySep Human Monocyte Enrichment Kit (STEMCELL Technologies, Vancouver, BC), as per manufacturer's instructions. For plastic adherence separation, PBMCs were incubated at 37°C for 2 h in serum-free RPMI. After incubation, unwanted cells were thoroughly washed from the adherent monocytes, which were then incubated in DC medium: RPMI supplemented with 10% defined FBS, 40 ng/ml recombinant human IL-4 and 50 ng/ml recombinant human GM-CSF (both ImmunoTools, Friesoythe, Germany). For immunomagnetic negative selection, PBMCs were incubated with magnetic particles coated with antibodies targeting unwanted cells (anti- CD2, CD3, CD16, CD19, CD20, CD56, CD66b, CD123 and glycophorin A). Labelled cells were magnetically separated and discarded, isolating the unlabelled monocytes. Monocytes were then incubated in DC medium. DCs were seeded on 24-, 48- or 96-well culture dishes at a density of 1 × 106 cells/ml and cultured for 6 days prior to infection with M. tuberculosis. The medium, containing fresh cytokines, was replaced every 2 to 3 days. Cytokines were also replenished 24 h after infection with M. tuberculosis, to maintain cytokine activity and DC phenotype throughout Mtb infection.
In vitro infection of DCs with M. tuberculosis
On the day of infection, mycobacteria were centrifuged at 3,800 rpm for 10 min and re-suspended in RPMI 1640 containing 10% defined FBS. Clumps were dispersed by passing the bacterial suspension through a 25 gauge needle eight times, and the sample was centrifuged at 800 rpm for 3 min to remove any remaining clumps. To determine the amount of Mtb necessary to achieve the required MOI, a CrystalSpec nephelometer (BD Diagnostic Systems, Sparks, MD) was used to estimate bacterial numbers in M. tuberculosis suspension. (Nephelometer bacterial number estimates was validated by counting colony-forming units (CFU) of bacterial suspension, plated on Middlebrook 7H10 agar plates, after 14 days). MOI were then calculated as bacteria per cell. DCs were infected at various MOI for 24 h, and extracellular bacteria were then removed by twice exchanging the medium with fresh DC medium. After 24 h infection, slides were prepared for acid-fast bacteria (AFB) staining to confirm phagocytosis. The cells were fixed for 10 min (H37Ra) or 24 h (H37Rv) in 2% paraformaldehyde (Sigma), applied to glass slides and left to air dry overnight. Slides were then stained with modified auramine O stain (Scientific Device Laboratory, Des Plaines, IL) for acid-fast bacteria. DC nuclei were counterstained with 10 μg of Hoechst 33358/ml (Sigma). The number of bacilli per cell was determined by observing the slides under an inverted fluorescence microscope (Olympus IX51, Olympus Corporation, Center Valley, PA). After infection, DCs were maintained in culture at 37°C for 1 to 3 days before harvesting.
Propidium iodide staining for IN Cell Analyzer viability assessment
Viability was assessed using the propidium iodide (PI) exclusion method for plasma membrane integrity of cells, and the nuclei were counterstained with Hoechst. Cells were incubated with 10 μg of PI/ml, Hoechst 33342 (10 μg/ml), and Hoechst 33358 (10 μg/ml) for 30 min at room temperature. The number of PI-positive cells relative to the total number of nuclei per field was counted by automated fluorescence microscopy using the IN Cell Analyzer 1000 and IN Cell Investigator software (GE Healthcare, Pittsburgh, PA). Each condition was assayed in triplicate, and 8 fields were counted in each well. Staurosporine (Sigma) (1 μM, diluted in serum-free RPMI) was applied for 24 h as a positive control for cell death.
DNA fragmentation (cell death ELISA)
The cell death detection ELISAPLUS kit (Roche Applied Science, Mannheim, Germany) was used to quantify M. tuberculosis-induced DNA fragmentation, as recommended by the manufacturer. Briefly, 1-3 days after infection, 48-well plates were centrifuged at 200 × g to sediment detached cells, the medium was discarded, and the cells were lysed. The lysate was subjected to antigen capture enzyme-linked immunosorbent assay (ELISA) to measure free nucleosomes, and the optical density at 405 nm (OD405) was read on a Victor2 plate reader (Wallac/Perkin Elmer, Waltham, MA). Triplicate wells were assayed for each condition. Staurosporine (Sigma) (1 μM, diluted in serum-free RPMI) was applied for 24 h as a positive control for DNA fragmentation.
The pan-caspase inhibitor, Q-VD-OPh (20 μM; Enzo Life Sciences AG, Lausen, Switzerland), was applied to DCs 4 h prior to infection with H37Ra and replenished every 24 h throughout the duration of infection
Caspase 3/7 activity was measured using the luminescent Caspase-Glo assay system (Promega, Madison, WI). DCs were cultured in 96-well plates and the assays were carried out in a total volume of 200 μl. After equilibration to room temperature, Caspase-Glo reagent was added to each well and gently mixed using a plate shaker at 300 rpm for 30 s. The plate was incubated at room temperature for 30 minutes and luminescence was then measured using a Victor2 plate reader.
Laser Scanning Confocal Microscopy
Following infection, DCs were fixed for 10 min (H37Ra) or 24 h (H37Rv) in 2% paraformaldehyde (Sigma), applied to glass slides and left to air dry overnight. The cells were then stained with modified auramine O stain for acid-fast bacteria and DC nuclei were counterstained with 10 μg/ml of Hoechst 33358. The slides were analysed using a Zeiss LSM 510 laser confocal microscope equipped with an Argon (488 nm excitation line; 510 nm emission detection) laser and a diode pulsed solid state laser (excitation 561 nm; emission 572 nm long pass filter) (Carl Zeiss MicroImaging GmbH, Oberkochen, Germany). Images were generated and viewed using LSM Image Browser (Carl Zeiss MicroImaging).
Dendritic cell surface markers were analysed by flow cytometry on a CyAn ADP flow cytometer (Dako/Beckman Coulter). Dendritic cells were infected with live H37Ra, or streptomycin-killed H37Ra at MOI 1 for 24 or 48 h. As a positive control for maturation, uninfected DCs were treated with LPS (Sigma; 1 μg/ml) for 24 h prior to staining for flow cytometry. Cells were incubated with antibodies for 30 min and fixed with 2% paraformaldehyde for at least 1 h prior to flow cytometry.
The following antibodies were used: FITC mouse anti-human CD14, FITC mouse IgG2a, κ isotype control, PE mouse anti-human CD209 (DC-SIGN), PE mouse IgG2b κ isotype control, APC mouse anti-human CD83, APC mouse IgG1 κ isotype control, FITC mouse anti-human CD86, FITC mouse IgG1 κ isotype control (all BD Pharmingen, San Diego, CA).
Cells were gated on forward scatter and side scatter to exclude clumps and debris. DCs were CD14- and DC-SIGN+ (constituting approximately 90% of gated cells). Results were analysed using Summit software version 4.3 (Dako/Beckman Coulter).
Dendritic cells were infected with live H37Ra or streptomycin-killed H37Ra at MOI 1 for 24 or 48 h. LPS was applied for 24 h (Sigma; 1 μg/ml) as a positive control for DC maturation and cytokine secretion. Cytokine secretion was measured in cell-free supernatants by ELISA using the Meso Scale Discovery SECTOR Imager 2400 and the following assays: human IL-6 assay, and human Th1/Th2 10-cytokine multiplex assay, capable of detecting IFN-γ, IL-1β, IL-10, IL-12p70, IL-13, IL-2, IL-4, IL-5, IL-8 and TNF-α (Meso Scale Discovery, Gaithersburg, MD). IL-4 measurements were disregarded, as DCs were maintained in culture with exogenous IL-4, rendering it impossible to distinguish levels secreted by the cells themselves.
Colony forming units and BacT/ALERT 3D
Dendritic cells were harvested 24 h or 72 h after infection with M. tuberculosis. Cells were centrifuged and washed 3 times at 800 rpm to remove extracellular bacteria. The cells were lysed in 0.1% Triton X-100 (Sigma) for 10 min. The resultant bacterial suspension was then passed through a 25 gauge needle eight times to disperse clumps. The bacilli were serially diluted x10-1 - x10-5 in Middlebrook 7H9 medium and plated on Middlebrook 7H10 agar (Difco) supplemented with oleic acid-albumin-dextrose-catalase (Becton Dickinson) and cycloheximide (Sigma), or inoculated into BacT/ALERT MP bottles (bioMérieux, Durham, NC). Agar plates were incubated at 37°C for 14 - 21 days and colony forming units were counted. BacT/ALERT MP bottles were incubated in a BacT/ALERT 3D automated microbial detection system (bioMérieux) and time to reach positivity was recorded, and a growth index was calculated, using the equation ((TTPDay1-TTPDay3)/TTPDay1)x100 as we have already published for the BD BACTEC liquid culture platform . In this equation TTP Day 1 is the time to culture positivity for infected DC lysates at Day 1, and TTP Day 3 is the time to positivity for infected DC lysates at Day 3.
Results are expressed as means ± the standard errors of the mean (SEM). The data were analyzed with GraphPad Prism 5 software (GraphPad Software, Inc., La Jolla, CA) statistical software using by repeated measures ANOVA with Tukey's post test, or (where stated) by the Friedman test followed by Dunns multiple comparison test. A P value of < 0.05 was considered statistically significant. Graphs were compiled using GraphPad Prism 5 software.
The authors wish to thank Dr Timothy Grant, Centre for Support and Analysis in Research (CSTAR) for providing advise on statistical analysis of the data. This work was supported by the Health Research Board of Ireland (HRB), Science Foundation Ireland (SFI-Immunology Cluster) and The Royal City of Dublin Hospital Trust. RCMR is a PhD candidate in the HRB-funded structured PhD programme in Molecular Medicine "From Genes to Function".
- WHO: Fact Sheet No 104: Tuberculosis. 2007, Geneva: World Health OrganisationGoogle Scholar
- WHO: Global Tuberculosis Control 2009: Epidemiology, Strategy, Financing. 2009, Geneva: World Health OrganisationGoogle Scholar
- Schreiber HA, Sandor M: The role of dendritic cells in mycobacterium-induced granulomas. Immunology Letters. 2010, 130 (1-2): 26-31. 10.1016/j.imlet.2009.12.009.PubMedPubMed CentralView ArticleGoogle Scholar
- Tascon RE, Soares CS, Ragno S, Stavropoulos E, Hirst EMA, Colston MJ: Mycobacterium tuberculosis-activated dendritic cells induce protective immunity in mice. Immunology. 2000, 99 (3): 473-480. 10.1046/j.1365-2567.2000.00963.x.PubMedPubMed CentralView ArticleGoogle Scholar
- Tian T, Woodworth J, Sköld M, Behar SM: In vivo depletion of CD11c+ cells delays the CD4+ T cell response to Mycobacterium tuberculosis and exacerbates the outcome of infection. The Journal of Immunology. 2005, 175 (5): 3268-3272.PubMedView ArticleGoogle Scholar
- Chackerian AA, Alt JM, Perera TV, Dascher CC, Behar SM: Dissemination of Mycobacterium tuberculosis is influenced by host factors and precedes the initiation of T-cell immunity. Infect Immun. 2002, 70 (8): 4501-4509. 10.1128/IAI.70.8.4501-4509.2002.PubMedPubMed CentralView ArticleGoogle Scholar
- Humphreys IR, Stewart GR, Turner DJ, Patel J, Karamanou D, Snelgrove RJ, Young DB: A role for dendritic cells in the dissemination of mycobacterial infection. Microbes and Infection. 2006, 8 (5): 1339-1346. 10.1016/j.micinf.2005.12.023.PubMedView ArticleGoogle Scholar
- Bigley V, Haniffa M, Doulatov S, Wang X-N, Dickinson R, McGovern N, Jardine L, Pagan S, Dimmick I, Chua I, Wallis J, Lordan J, Morgan C, Kumararatne DS, Doffinger R, van der Burg M, van Dongen J, Cant A, Dick JE, Hambleton S, Collin M: The human syndrome of dendritic cell, monocyte, B and NK lymphoid deficiency. The Journal of Experimental Medicine. 2011, 208 (2): 227-234. 10.1084/jem.20101459.PubMedPubMed CentralView ArticleGoogle Scholar
- Hambleton S, Salem S, Bustamante J, Bigley V, Boisson-Dupuis S, Azevedo J, Fortin A, Haniffa M, Ceron-Gutierrez L, Bacon CM, et al: IRF8 mutations and human dendritic-cell immunodeficiency. New England Journal of Medicine. 2011, 365: 127-138. 10.1056/NEJMoa1100066.PubMedPubMed CentralView ArticleGoogle Scholar
- O'Sullivan MP, O'Leary S, Kelly DM, Keane J: A caspase-independent pathway mediates macrophage cell death in response to Mycobacterium tuberculosis infection. Infect Immun. 2007, 75 (4): 1984-1993. 10.1128/IAI.01107-06.PubMedPubMed CentralView ArticleGoogle Scholar
- Rodrigues MF, Barsante MM, Alves CCS, Souza MA, Ferreira AP, Amarante-Mendes GP, Teixeira HC: Apoptosis of macrophages during pulmonary Mycobacterium bovis infection: correlation with intracellular bacillary load and cytokine levels. Immunology. 2009, 128 (1pt2): e691-e699. 10.1111/j.1365-2567.2009.03062.x.PubMedPubMed CentralView ArticleGoogle Scholar
- Sohn H, Lee KS, Kim SY, Shin DM, Shin SJ, Jo EK, Park JK, Kim HJ: Induction of cell death in human macrophages by a highly virulent Korean isolate of Mycobacterium tuberculosis and the virulent strain H37Rv. Scandinavian Journal of Immunology. 2009, 69 (1): 43-50. 10.1111/j.1365-3083.2008.02188.x.PubMedView ArticleGoogle Scholar
- Keane J, Balcewicz-Sablinska MK, Remold HG, Chupp GL, Meek BB, Fenton MJ, Kornfeld H: Infection by Mycobacterium tuberculosis promotes human alveolar macrophage apoptosis. Infect Immun. 1997, 65 (1): 298-304.PubMedPubMed CentralGoogle Scholar
- Schaible UE, Winau F, Sieling PA, Fischer K, Collins HL, Hagens K, Modlin RL, Brinkmann V, Kaufmann SHE: Apoptosis facilitates antigen presentation to T lymphocytes through MHC-I and CD1 in tuberculosis. Nat Med. 2003, 9 (8): 1039-1046. 10.1038/nm906.PubMedView ArticleGoogle Scholar
- Winau F, Weber S, Sad S, de Diego J, Hoops SL, Breiden B, Sandhoff K, Brinkmann V, Kaufmann SHE, Schaible UE: Apoptotic vesicles crossprime CD8 T cells and protect against tuberculosis. Immunity. 2006, 24 (1): 105-117. 10.1016/j.immuni.2005.12.001.PubMedView ArticleGoogle Scholar
- Montes-Worboys A, Brown S, Regev D, Bellew BF, Mohammed KA, Faruqi I, Sharma P, Moudgil B, Antony VB: Targeted delivery of amikacin into granuloma. Am J Respir Crit Care Med. 2010, 182 (12): 1546-1553. 10.1164/rccm.200912-1888OC.PubMedPubMed CentralView ArticleGoogle Scholar
- McShane H, Behboudi S, Goonetilleke N, Brookes R, Hill AVS: Protective immunity against Mycobacterium tuberculosis induced by dendritic cells pulsed with both CD8+- and CD4+-T-cell epitopes from antigen 85A. Infect Immun. 2002, 70 (3): 1623-1626. 10.1128/IAI.70.3.1623-1626.2002.PubMedPubMed CentralView ArticleGoogle Scholar
- Badovinac VP, Messingham KAN, Jabbari A, Haring JS, Harty JT: Accelerated CD8+ T-cell memory and prime-boost response after dendritic-cell vaccination. Nat Med. 2005, 11 (7): 748-756. 10.1038/nm1257.PubMedView ArticleGoogle Scholar
- Kong CU, Ng LG, Nambiar JK, Spratt JM, Weninger W, Triccas JA: Targeted induction of antigen expression within dendritic cells modulates antigen-specific immunity afforded by recombinant BCG. Vaccine. 2011, 29 (7): 1374-1381. 10.1016/j.vaccine.2010.12.070.PubMedView ArticleGoogle Scholar
- Kerr JFR, Wyllie AH, Currie AR: Apoptosis: A basic biological phenomenon with wide-ranging implications in tissue kinetics. British Journal of Cancer. 1972, 26 (4): 239-257. 10.1038/bjc.1972.33.PubMedPubMed CentralView ArticleGoogle Scholar
- Mevorach D, Trahtemberg U, Krispin A, Attalah M, Zazoun J, Tabib A, Grau A, Verbovetski-Reiner I: What do we mean when we write "senescence,""apoptosis,""necrosis," or "clearance of dying cells"?. Annals of the New York Academy of Sciences. 2010, 1209 (1): 1-9. 10.1111/j.1749-6632.2010.05774.x.PubMedView ArticleGoogle Scholar
- Kroemer G, Galluzzi L, Vandenabeele P, Abrams J, Alnemri ES, Baehrecke EH, Blagosklonny MV, El-Deiry WS, Golstein P, Green DR, Hengartner M, Knight RA, Kumar S, Lipton SA, Malorni W, Nuñez G, Peter ME, Tschopp J, Yuan J, Piacentini M, Zhivotovsky B, Melino G: Classification of cell death: recommendations of the Nomenclature Committee on Cell Death 2009. Cell Death Differentiation. 2009, 16 (1): 3-11. 10.1038/cdd.2008.150.PubMedPubMed CentralView ArticleGoogle Scholar
- Caserta TM, Smith AN, Gultice AD, Reedy MA, Brown TL: Q-VD-OPh, a broad spectrum caspase inhibitor with potent antiapoptotic properties. Apoptosis. 2003, 8 (4): 345-352. 10.1023/A:1024116916932.PubMedView ArticleGoogle Scholar
- Kroemer G, Martin SJ: Caspase-independent cell death. Nat Med. 2005, 11 (7): 725-730. 10.1038/nm1263.PubMedView ArticleGoogle Scholar
- Molloy A, Laochumroonvorapong P, Kaplan G: Apoptosis, but not necrosis, of infected monocytes is coupled with killing of intracellular bacillus Calmette-Guérin. The Journal of Experimental Medicine. 1994, 180 (4): 1499-1509. 10.1084/jem.180.4.1499.PubMedView ArticleGoogle Scholar
- Laochumroonvorapong P, Paul S, Elkon K, Kaplan G: H2O2 induces monocyte apoptosis and reduces viability of Mycobacterium avium-M. intracellulare within cultured human monocytes. Infect Immun. 1996, 64 (2): 452-459.PubMedPubMed CentralGoogle Scholar
- Alemán M, de la Barrera S, Schierloh P, Yokobori N, Baldini M, Musella R, Abbate E, Sasiain M: Spontaneous or Mycobacterium tuberculosis-induced apoptotic neutrophils exert opposite effects on the dendritic cell-mediated immune response. European Journal of Immunology. 2007, 37 (6): 1524-1537. 10.1002/eji.200636771.PubMedView ArticleGoogle Scholar
- Hedlund S, Persson A, Vujic A, Che KF, Stendahl O, Larsson M: Dendritic cell activation by sensing Mycobacterium tuberculosis-induced apoptotic neutrophils via DC-SIGN. Human Immunology. 2010, 71 (6): 535-540. 10.1016/j.humimm.2010.02.022.PubMedView ArticleGoogle Scholar
- Persson YAZ, Blomgran-Julinder R, Rahman S, Zheng L, Stendahl O: Mycobacterium tuberculosis-induced apoptotic neutrophils trigger a pro-inflammatory response in macrophages through release of heat shock protein 72, acting in synergy with the bacteria. Microbes and Infection. 2008, 10 (3): 233-240. 10.1016/j.micinf.2007.11.007.PubMedView ArticleGoogle Scholar
- Keane J, Remold HG, Kornfeld H: Virulent Mycobacterium tuberculosis strains evade apoptosis of infected alveolar macrophages. J Immunol. 2000, 164 (4): 2016-2020.PubMedView ArticleGoogle Scholar
- Oddo M, Renno T, Attinger A, Bakker T, MacDonald HR, Meylan PRA: Fas ligand-induced apoptosis of infected human macrophages reduces the viability of intracellular Mycobacterium tuberculosis. J Immunol. 1998, 160 (11): 5448-5454.PubMedGoogle Scholar
- Randhawa AK, Ziltener HJ, Stokes RW: CD43 controls the intracellular growth of Mycobacterium tuberculosis through the induction of TNF-α-mediated apoptosis. Cellular Microbiology. 2008, 10 (10): 2105-2117. 10.1111/j.1462-5822.2008.01194.x.PubMedView ArticleGoogle Scholar
- Nogueira CV, Lindsten T, Jamieson AM, Case CL, Shin S, Thompson CB, Roy CR: Rapid pathogen-induced apoptosis: a mechanism used by dendritic cells to limit intracellular replication of Legionella pneumophila. PLoS Pathog. 2009, 5 (6): e1000478-10.1371/journal.ppat.1000478.PubMedPubMed CentralView ArticleGoogle Scholar
- Fortsch D, Rollinghoff M, Stenger S: IL-10 converts human dendritic cells into macrophage-like cells with increased antibacterial activity against virulent Mycobacterium tuberculosis. J Immunol. 2000, 165 (2): 978-987.PubMedView ArticleGoogle Scholar
- Bodnar KA, Serbina NV, Flynn JL: Fate of Mycobacterium tuberculosis within murine dendritic cells. Infect Immun. 2001, 69 (2): 800-809. 10.1128/IAI.69.2.800-809.2001.PubMedPubMed CentralView ArticleGoogle Scholar
- Tailleux L, Neyrolles O, Honore-Bouakline S, Perret E, Sanchez F, Abastado J-P, Lagrange PH, Gluckman JC, Rosenzwajg M, Herrmann J-L: Constrained intracellular survival of Mycobacterium tuberculosis in human dendritic cells. J Immunol. 2003, 170 (4): 1939-1948.PubMedView ArticleGoogle Scholar
- Winau F, Hegasy G, Kaufmann SHE, Schaible UE: No life without death--apoptosis as prerequisite for T cell activation. Apoptosis. 2005, 10 (4): 707-715. 10.1007/s10495-005-2940-6.PubMedView ArticleGoogle Scholar
- Dieli F, Troye-Blomberg M, Ivanyi J, Fournié JJ, Krensky AM, Bonneville M, Peyrat MA, Caccamo N, Sireci G, Salerno A: Granulysin-dependent killing of intracellular and extracellular Mycobacterium tuberculosis by Vγ9/Vδ2 T lymphocytes. Journal of Infectious Diseases. 2001, 184 (8): 1082-1085. 10.1086/323600.PubMedView ArticleGoogle Scholar
- Canaday DH, Wilkinson RJ, Li Q, Harding CV, Silver RF, Boom WH: CD4+ and CD8+ T cells kill intracellular Mycobacterium tuberculosis by a perforin and Fas/Fas ligand-independent mechanism. The Journal of Immunology. 2001, 167 (5): 2734-2742.PubMedView ArticleGoogle Scholar
- Bastian M, Braun T, Bruns H, Röllinghoff M, Stenger S: Mycobacterial lipopeptides elicit CD4+ CTLs in Mycobacterium tuberculosis-infected humans. The Journal of Immunology. 2008, 180 (5): 3436-3446.PubMedView ArticleGoogle Scholar
- Martino A, Casetti R, Sacchi A, Poccia F: Central memory Vγ9Vδ2 T lymphocytes primed and expanded by Bacillus Calmette-Guérin-infected dendritic cells kill mycobacterial-infected monocytes. The Journal of Immunology. 2007, 179 (5): 3057-3064.PubMedView ArticleGoogle Scholar
- Fernandes-Alnemri T, Wu J, Yu JW, Datta P, Miller B, Jankowski W, Rosenberg S, Zhang J, Alnemri ES: The pyroptosome: a supramolecular assembly of ASC dimers mediating inflammatory cell death via caspase-1 activation. Cell Death Differ. 2007, 14 (9): 1590-1604. 10.1038/sj.cdd.4402194.PubMedPubMed CentralView ArticleGoogle Scholar
- Green DR, Ferguson T, Zitvogel L, Kroemer G: Immunogenic and tolerogenic cell death. Nat Rev Immunol. 2009, 9 (5): 353-363. 10.1038/nri2545.PubMedPubMed CentralView ArticleGoogle Scholar
- Torchinsky MB, Garaude J, Blander JM: Infection and apoptosis as a combined inflammatory trigger. Current Opinion in Immunology. 2010, 22 (1): 55-62. 10.1016/j.coi.2010.01.003.PubMedView ArticleGoogle Scholar
- Briken V, Miller JL: Living on the edge: inhibition of host cell apoptosis by Mycobacterium tuberculosis. Future Microbiology. 2008, 3 (4): 415-422. 10.2217/174609220.127.116.115.PubMedPubMed CentralView ArticleGoogle Scholar
- Ordway D, Henao-Tamayo M, Orme IM, Gonzalez-Juarrero M: Foamy macrophages within lung granulomas of mice infected with Mycobacterium tuberculosis express molecules characteristic of dendritic cells and antiapoptotic markers of the TNF receptor-associated factor family. The Journal of Immunology. 2005, 175 (6): 3873-3881.PubMedView ArticleGoogle Scholar
- Zheng H, Lu L, Wang B, Pu S, Zhang X, Zhu G, Shi W, Zhang L, Wang H, Wang S, Zhao G, Zhang Y: Genetic basis of virulence attenuation revealed by comparative genomic analysis of Mycobacterium tuberculosis strain H37Ra versus H37Rv. PLoS One. 2008, 3 (6): e2375-10.1371/journal.pone.0002375.PubMedPubMed CentralView ArticleGoogle Scholar
- Li AH, Waddell SJ, Hinds J, Malloff CA, Bains M, Hancock RE, Lam WL, Butcher PD, Stokes RW: Contrasting transcriptional responses of a virulent and an attenuated strain of Mycobacterium tuberculosis infecting macrophages. PLoS One. 2010, 5 (6): e11066-10.1371/journal.pone.0011066.PubMedPubMed CentralView ArticleGoogle Scholar
- Frigui W, Bottai D, Majlessi L, Monot M, Josselin E, Brodin P, Garnier T, Gicquel B, Martin C, Leclerc C, Cole ST, Brosch R: Control of M. tuberculosis ESAT-6 secretion and specific T cell recognition by PhoP. PLoS Pathog. 2008, 4 (2): e33-10.1371/journal.ppat.0040033.PubMedPubMed CentralView ArticleGoogle Scholar
- Silver RF, Li Q, Ellner JJ: Expression of virulence of Mycobacterium tuberculosis within human monocytes: virulence correlates with intracellular growth and induction of tumor necrosis factor alpha but not with evasion of lymphocyte-dependent monocyte effector functions. Infect Immun. 1998, 66 (3): 1190-1199.PubMedPubMed CentralGoogle Scholar
- Zhang M, Gong J, Lin Y, Barnes PF: Growth of virulent and avirulent Mycobacterium tuberculosis strains in human macrophages. Infect Immun. 1998, 66 (2): 794-799.PubMedPubMed CentralGoogle Scholar
- Nigou J, Vercellone A, Puzo G: New structural insights into the molecular deciphering of mycobacterial lipoglycan binding to C-type lectins: lipoarabinomannan glycoform characterization and quantification by capillary electrophoresis at the subnanomole level. J Mol Biol. 2000, 299 (5): 1353-1362. 10.1006/jmbi.2000.3821.PubMedView ArticleGoogle Scholar
- Barry DP, Beaman BL: Modulation of eukaryotic cell apoptosis by members of the bacterial order Actinomycetales. Apoptosis. 2006, 11 (10): 1695-1707. 10.1007/s10495-006-9236-3.PubMedView ArticleGoogle Scholar
- Welin A, Eklund D, Stendahl O, Lerm M: Human macrophages infected with a high burden of ESAT-6-expressing M. tuberculosis undergo caspase-1- and cathepsin B-independent necrosis. PLoS One. 2011, 6 (5): e20302-10.1371/journal.pone.0020302.PubMedPubMed CentralView ArticleGoogle Scholar
- Obeid M, Tesniere A, Ghiringhelli F, Fimia GM, Apetoh L, Perfettini JL, Castedo M, Mignot G, Panaretakis T, Casares N, Métivier D, Larochette N, van Endert P, Ciccosanti F, Piacentini M, Zitvogel L, Kroemer G: Calreticulin exposure dictates the immunogenicity of cancer cell death. Nat Med. 2007, 13 (1): 54-61. 10.1038/nm1523.PubMedView ArticleGoogle Scholar
- Kazama H, Ricci J-E, Herndon JM, Hoppe G, Green DR, Ferguson TA: Induction of immunological tolerance by apoptotic cells requires caspase-dependent oxidation of high-mobility group box-1 protein. Immunity. 2008, 29 (1): 21-32. 10.1016/j.immuni.2008.05.013.PubMedPubMed CentralView ArticleGoogle Scholar
- Pang B, Neijssen J, Qiao X, Janssen L, Janssen H, Lippuner C, Neefjes J: Direct antigen presentation and gap junction mediated cross-presentation during apoptosis. The Journal of Immunology. 2009, 183 (2): 1083-1090. 10.4049/jimmunol.0900861.PubMedView ArticleGoogle Scholar
- Wolf AJ, Linas B, Trevejo-Nunez GJ, Kincaid E, Tamura T, Takatsu K, Ernst JD: Mycobacterium tuberculosis infects dendritic cells with high frequency and impairs their function in vivo. J Immunol. 2007, 179 (4): 2509-2519.PubMedView ArticleGoogle Scholar
- Wolf AJ, Desvignes L, Linas B, Banaiee N, Tamura T, Takatsu K, Ernst JD: Initiation of the adaptive immune response to Mycobacterium tuberculosis depends on antigen production in the local lymph node, not the lungs. J Exp Med. 2008, 205 (1): 105-115. 10.1084/jem.20071367.PubMedPubMed CentralView ArticleGoogle Scholar
- Khader SA, Partida-Sanchez S, Bell G, Jelley-Gibbs DM, Swain S, Pearl JE, Ghilardi N, deSauvage FJ, Lund FE, Cooper AM: Interleukin 12p40 is required for dendritic cell migration and T cell priming after Mycobacterium tuberculosis infection. J Exp Med. 2006, 203 (7): 1805-1815. 10.1084/jem.20052545.PubMedPubMed CentralView ArticleGoogle Scholar
- Alaniz RC, Sandall S, Thomas EK, Wilson CB: Increased dendritic cell numbers impair protective immunity to intracellular bacteria despite augmenting antigen-specific CD8+ T lymphocyte responses. J Immunol. 2004, 172 (6): 3725-3735.PubMedView ArticleGoogle Scholar
- Remoli ME, Giacomini E, Petruccioli E, Gafa V, Severa M, Gagliardi MC, Iona E, Pine R, Nisini R, Coccia EM: Bystander inhibition of dendritic cell differentiation by Mycobacterium tuberculosis-induced IL-10. Immunol Cell Biol. 2010Google Scholar
- Floto RA, MacAry PA, Boname JM, Mien TS, Kampmann B, Hair JR, Huey OS, Houben ENG, Pieters J, Day C, Oehlmann W, Singh M, Smith KG, Lehner PJ: Dendritic cell stimulation by mycobacterial Hsp70 is mediated through CCR5. Science. 2006, 314 (5798): 454-458. 10.1126/science.1133515.PubMedView ArticleGoogle Scholar
- Rajashree P, Supriya P, Das SD: Differential migration of human monocyte-derived dendritic cells after infection with prevalent clinical strains of Mycobacterium tuberculosis. Immunobiology. 2008, 213 (7): 567-575. 10.1016/j.imbio.2008.01.007.PubMedView ArticleGoogle Scholar
- Bansal K, Sinha AY, Ghorpade DS, Togarsimalemath SK, Patil SA, Kaveri SV, Balaji KN, Bayry J: Src homology 3-interacting domain of Rv1917c of Mycobacterium tuberculosis induces selective maturation of human dendritic cells by regulating PI3K-MAPK-NF-κB signaling and drives Th2 immune responses. Journal of Biological Chemistry. 2010, 285 (47): 36511-36522. 10.1074/jbc.M110.158055.PubMedPubMed CentralView ArticleGoogle Scholar
- Wang C, Peyron P, Mestre O, Kaplan G, van Soolingen D, Gao Q, Gicquel B, Neyrolles O: Innate immune response to Mycobacterium tuberculosis Beijing and other genotypes. PLoS ONE. 2010, 5 (10): e13594-10.1371/journal.pone.0013594.PubMedPubMed CentralView ArticleGoogle Scholar
- Torchinsky MB, Garaude J, Martin AP, Blander JM: Innate immune recognition of infected apoptotic cells directs TH17 cell differentiation. Nature. 2009, 458 (7234): 78-82. 10.1038/nature07781.PubMedView ArticleGoogle Scholar
- Nakano H, Nagata T, Suda T, Tanaka T, Aoshi T, Uchijima M, Kuwayama S, Kanamaru N, Chida K, Nakamura H, Okada M, Koide Y: Immunization with dendritic cells retrovirally transduced with mycobacterial antigen 85A gene elicits the specific cellular immunity including cytotoxic T-lymphocyte activity specific to an epitope on antigen 85A. Vaccine. 2006, 24 (12): 2110-2119. 10.1016/j.vaccine.2005.11.026.PubMedView ArticleGoogle Scholar
- Keane J, Shurtleff B, Kornfeld H: TNF-dependent BALB/c murine macrophage apoptosis following Mycobacterium tuberculosis infection inhibits bacillary growth in an IFN-γ independent manner. Tuberculosis. 2002, 82 (2-3): 55-61. 10.1054/tube.2002.0322.PubMedView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.