Skip to main content
NCBI home page
The Journal of Infectious Diseases logoLink to The Journal of Infectious Diseases
. 2018 Mar 14;218(10):1653–1662. doi: 10.1093/infdis/jiy142

Limited Antimycobacterial Efficacy of Epitope Peptide Administration Despite Enhanced Antigen-Specific CD4 T-Cell Activation

Joel D Ernst 1,2,3,✉, Amber Cornelius 1, Ludovic Desvignes 1, Jacqueline Tavs 1, Brian A Norris 1
PMCID: PMC6173573  PMID: 29548008

Systemic administration of T cell epitope peptides increased activation of antigen-specific T cells in Mycobacterium tuberculosis-infected mice, but the antimycobacterial effect was modest. Additional work indicated that CD4 T cells are segregated from M. tuberculosis-infected cells in the lungs.

Keywords: CD4 T cell, epitope, peptide:MHC tetramers, T-cell activation, tuberculosis

Abstract

Background

Infection with Mycobacterium tuberculosis is associated with inconsistent and incomplete elimination of the bacteria, despite development of antigen-specific T-cell responses. One mechanism used by M tuberculosis is to limit availability of antigen for activation of CD4 T cells.

Methods

We examined the utility of systemic administration of epitope peptides to activate pre-existing T cells in mice infected with M tuberculosis.

Results

We found that systemic peptide administration (1) selectively activates T cells specific for the epitope peptide, (2) loads major histocompatibility complex class II on lung macrophages and dendritic cells, (3) activates CD4 T cells in the lung parenchyma, (4) and has little antimycobacterial activity.

Conclusions

Further studies revealed that CD4 T cells in lung lesions are distant from the infected cells, suggesting that, even if they can be activated, the positioning of CD4 T cells and their direct interactions with infected cells may be limiting determinants of immunity in tuberculosis.

Immunologic control of Mycobacterium tuberculosis requires CD4 T cells [1, 2] that recognize specific antigens and become activated to perform effector functions [3, 4]. In most individuals, immune responses limit disease progression. However, in many humans and in mouse models, T cells do not eliminate the bacteria, indicating that M tuberculosis limits or evades T-cell responses. One evasion mechanism is to limit antigen availability, causing suboptimal T-cell activation at the site of infection [5–7]. Antigen availability may be limited by reduced antigen gene expression [5], inefficient antigen processing [8] or presentation [9], or other mechanisms [10]. Because immune control of M tuberculosis depends on direct recognition of infected cells by CD4 T cells [11], any mechanism that limits this cell-cell interaction could limit the efficacy of CD4 T-cell responses.

We previously reported that limited antigen availability can be overcome by systemic administration of an M tuberculosis epitope peptide, Ag85B240-254 (peptide 25), which activated endogenous CD4 effector T cells and Ag85B240-254-specific T-cell receptor (TCR) transgenic CD4 T cells and modestly reduced M tuberculosis burdens [5]. To further understand the action of peptide 25 in M tuberculosis-infected mice, we characterized the responding T cells and their mechanisms of activation, and we examined the effects of additional epitope peptides. We found that (1) activation of T cells is epitope-specific, (2) systemic administration loads antigen-presenting cells with peptide, and (3) responding T cells are in the lung parenchyma. Furthermore, epitope peptides from other antigens activate CD4 T cells, but their antimycobacterial activity does not exceed that of peptide 25. In further studies of the discordance between T-cell activation and antimycobacterial efficacy, we found that M tuberculosis-infected cells in the lungs are sequestered from CD4 T cells, suggesting that spatial separation of CD4 T cells and infected cells contributes to bacillary persistence.

METHODS

Mice

C57BL/6 and CB6F1/J mice were from Taconic Farms and The Jackson Laboratory, respectively. P25TCR-Tg mice [12–14] were bred in the New York University (NYU) School of Medicine animal facility. All animal procedures were approved by the NYU School of Medicine Institutional Animal Care and Use Committee.

Aerosol Infection

Eight-week-old male and female mice were infected with ~100 colony-forming units (CFUs) of M tuberculosis H37Rv or H37Rv-EGFP via aerosol [14]. To confirm the inoculum, 24 hours postinfection, lungs from 4 mice were homogenized and plated on 7H11 agar; colonies were counted 3 weeks later.

Administration of Synthetic Peptides

Mice were treated intravenously with synthetic peptides (EZBiolab) representing the following epitopes (the sequence of the peptide precedes the semicolon, and the restricting element or haplotype follows, when known): Ag85B240-254 (amino acid numbering of the mature protein) (FQDAYNAAGGHNAVF; I-Ab), Ag85A104-123 (LTSELPGWLQANRHVKPTGS; I-Ed), Ag85A261-280 (THSWEYWGAQLNAMKPDLQR; I-Ab), Mtb32A309-318 (GAPINSATAM; H2Db), HspX21-40 (LFAAFPSFAGLRPTFDTRLM; I-Ab), HspX71-91 (RDGQLTIKAERTEQKDFDGR; I-Ad or I-Ed), EsxH/TB10.421–38 (YAGTLQSLGAEIAVEQAA; H-2b), EsxH/TB10.461–78 (AMEDLVRAYHAMSSTHEA; H-2b or H2d), EsxH/TB10.471–88 (AMSSTHEANTMAMMARDT), ESAT-61-20 (MTEQQWNFAGIEAAASAIQG; I-Ab), or OVA323-339 (ISQAVHAAHAEINEAGR; I-Ab). Peptides were >95% pure and administered as 50 μg of peptide in 200 µL sterile phosphate-buffered saline (PBS).

Tissue Processing for Flow Cytometry and Bacterial Burden Quantification

After euthanasia, lungs were removed, diced, and digested, and single-cell suspensions were prepared [14]. Single-cell suspensions were incubated with control (hCLIP:I-Ab) or M tuberculosis-specific major histocompatibility complex class (MHC) II tetramers (Ag85B280-294:I-Ab, ESAT64-17:I-Ab, or EsxG46-61:I-Ab) for 2–3 hours at 37°C in complete Roswell Park Memorial Institute medium + 5% fetal calf serum. Cells were stained with Zombie-NIR in PBS, then with labeled monoclonal antibodies in PBS + 3% bovine serum albumin. CD11b-Brilliant Violet 711 (clone M1/70), CD3-fluorescein isothiocyanate (145-2C11), CD4-Brilliant Violet 421 (GK1.5), and CD8α-Brilliant Violet 605 (53–6.7) were from BioLegend; Siglec-F-APC-R700 (E50-2440) was from BD. Surface stained cells were permeabilized with cytofix/cytoperm (BD) and stained for intracellular interferon (IFN)γ with Alexa Fluor 647 (XMG1.2) antibody.

Tetramers were from the National Institute of Allergy and Infectious Diseases Tetramer Core Facility. The Ag85B tetramer nomenclature (I-Ab Ag85B280-294) counts the 40 amino acid signal sequence; the epitope peptide in this tetramer is identical to that recognized by P25TCR-Tg CD4 T cells, where it is termed Ag85B240-254. For detection of IFNγ-producing cells without ex vivo restimulation, lungs were processed 6 hours after the peptide dose, with all cell isolation steps except collagenase/DNase digestion and ammonium-chloride-potassium (ACK) lysis done on ice. Data were acquired using an LSR II flow cytometer (BD) and analyzed using FlowJo. Bacteria were quantitated in lung cell suspensions taken before ACK lysis, diluted in PBS/0.5% Tween 80, and plated on 7H11 agar. Colonies were counted after 3 weeks at 37°C.

P25TCR-Tg Th1 Effector T Cells

P25TCR-Tg Th1 effector T cells were prepared as described [5]. For adoptive transfer, 3 × 106 cells were injected 25 days postinfection, isolated from lungs 48 hours later, and analyzed by flow cytometry.

Macrophage and Dendritic Cell Isolation and Analysis of In Vivo Peptide Loading of Major Histocompatibility Complex Class II

Mice infected with M tuberculosis H37Rv-enhanced green fluorescent protein (EGFP) received peptide 25 or ovalbumin (OVA); 1 hour later, lung cells were isolated, pooled (3 mice/pool), and stained with antibodies to CD11c, CD11b, and Gr-1 [14]. Stained cells were sorted using a BSL3-contained Sony Synergy flow sorter [15, 16], then incubated with P25TCR-Tg Th1 cells (10 T cells/antigen-presenting cell) in V-bottom 96-well plates. After 48 hours, IFNγ was quantitated by enzyme-linked immunosorbent assay (BD).

Definition of Intravascular and Parenchymal T Cells

P25TCR-Tg Th1 (CD45.1+) cells were transferred into C57BL/6 mice 26 days postinfection. Forty-eight hours later, mice received peptide 25 or OVA peptide, and 6 hours later they received pan CD45-PE-Cy7 (2.5 µg) intravenously. Three minutes later, mice were euthanized and lung cells were stained for CD45.1, CD4, Thy1.2, and CD11b, then fixed, permeabilized, and stained for IFNγ. Cells accessible to the injected antibody (CD45-PE-Cy7+) were defined as intravascular, and cells shielded from the intravenous antibody (CD45-PE-Cy7−) were defined as parenchymal [17–19]. CD45.1 was used to distinguish P25TCRTg cells (CD45.1+) from endogenous cells (CD45.1−).

Tissue Section Preparation, Staining, and Analysis

Mice were euthanized 4 or 8 weeks after H37Rv-EGFP infection. Their lungs were perfused intratracheally with Optimal Cutting Temperature (OCT) medium before being removed, embedded in OCT, and frozen with liquid nitrogen. Cryostat sections (6 μM) were transferred to slides and fixed in ice-cold acetone (15 minutes). For staining, sections were thawed, rehydrated in PBS (20 minutes), blocked with PBS + 10% fetal bovine serum, and stained with CD4-Alexa Fluor 647 (GK1.5), CD11b-Alexa Fluor 594 (M1/70), and FcR blocking antibody (2.4G2) for 2 hours. After washing with PBS and staining with Hoescht nuclear dye, images were acquired on a Nikon Eclipse Ti epifluorescence microscope. Multiple images were taken using an oil-immersion ×60 objective with multiple z-stacks before images were tiled together and deconvoluted using NIS-Elements software to reduce out-of-focus fluorescence.

Images were first converted from ND2 to TIFF, then color balance adjusted to reduce background fluorescence, using ImageJ. To measure distances between bacilli and CD4 T cells, we wrote a program in Jython that automatically applied Otsu’s threshold, watershed, and fill holes. The program then utilized the ImageJ particle analyzer function to identify all CD4 T cells and all EGFP-expressing bacilli. The pixel coordinates of bacilli and CD4 T-cell nuclei were identified, and the Euclidean distance between them was calculated. The distance between the nearest CD4 T cell and each bacillus (representing an infected cell) was then tabulated, compiled into a table, and graphed using R and GraphPad Prism. Estimates of average myeloid cell size were obtained by first applying threshold and fill holes to the CD11b and Hoescht nuclear stain signals in ImageJ. The total area of CD11b signal was then determined, and nuclei within CD11b+ cells was enumerated. The average area per CD11b+ cell was then determined, and the diameter of a theoretical circular CD11b cell was calculated.

Statistical Analyses

Student’s t test, one-way analysis of variance (ANOVA) with Dunnett’s posttest, or 2-way ANOVA with Tukey’s posttest were performed as specified in the figure legends, using Prism (GraphPad): *, P < .05; **, P < .005; ***, P < .001; ****, P < .0001; n.s. = not significant.

RESULTS

Systemic Epitope Peptide Administration Selectively Activates Epitope-Specific T Cells In Vivo

Previous studies demonstrated that administration of peptide 25 (Ag85B240-254) activated peptide 25-specific P25TCRTg CD4 T cells in vivo, and it also activated a fraction of the endogenous CD4 T cells in M tuberculosis-infected mice. Because those studies did not evaluate whether the responding endogenous T cells were specific, we used peptide:MHC tetramers to identify CD4 T cells specific for I:Ab-Ag85B280-294, I:Ab-ESAT-64-17, or I:Ab-EsxG46-61 after injection of epitope peptides OVA323-339, Ag85B240-254, or ESAT-63-15, with activation of tetramer-positive cells indicated by intracellular IFNγ expression. Other intracellular cytokines (interleukin-17, tumor necrosis factor) were below the limit of detection without ex vivo restimulation, so we used IFNγ to indicate T-cell activation. Approximately 9% of CD4+Ag85B tetramer+ cells were IFNγ+ after control (OVA323-339) peptide injection (Figure 1A and B); CD4+Ag85B tetramer+IFNγ+ T cells increased ~2.5 fold with injection of Ag85B peptide but not ESAT-6 peptide. Similarly, 3%–5% of ESAT-6 tetramer+ cells were IFNγ+ after injection of control peptide; this increased markedly (to ~23%) after injection of ESAT-6 peptide but not Ag85B peptide. Neither the Ag85B nor the ESAT-6 peptide activated EsxG tetramer-specific CD4 T cells. These results demonstrate that systemic injection of M tuberculosis epitope peptides activates T cells specific for that epitope and not T cells specific for other antigens.

Figure 1.

Figure 1.

Open in a new tab

In vivo peptide-induced activation of CD4 T cells is epitope-specific. C57BL/6 mice were infected with Mycobacterium tuberculosis H37Rv by aerosol; on day 28 postinfection, peptides (O = OVA323-339; 85B = Ag85B240-254; E = ESAT-63-15) were injected intravenously, and lungs were harvested 6 hours later. Single-cell suspensions were stained with antibodies to CD4 and with the indicated I:Ab tetramers, then cells were fixed and permeabilized, stained for intracellular interferon (IFN)γ, and the frequencies of CD4+tetramer+IFNγ+ cells were quantitated by flow cytometry. (A) Representative flow cytometry plots. (B) Summary data; each symbol represents data from an individual mouse. Note that the amino acid numbering of the injected Ag85B peptide (240–254) and the Ag85B tetramer (280–294) differ because the latter includes the 40-residue signal sequence, whereas the former does not; the identities and sequences of the peptides are identical. P values (**, <.01; ***, <.001) were obtained by 2-way analysis of variance with Tukey’s posttest, and they compare the values obtained with each of the designated M tuberculosis peptides, compared with the values obtained with the ovalbumin peptide, for each set of tetramer-specific cells.

Epitope Peptide-Responsive T Cells Are in the Lung Parenchyma

Recent studies revealed that in M tuberculosis-infected mice, CD4 T cells can be in the lung vasculature or parenchyma, and that only T cells in the parenchyma contribute to restricting bacterial growth [18, 20–22]. Therefore, we examined whether epitope peptide injection activates lung intravascular or parenchymal T cells. Mycobacterium tuberculosis-infected mice received Ag85B-specific P25TCR-Tg Th1 cells 27 days postinfection; on day 28, they received Ag85B240-254 or OVA323-339 (control) peptide, followed 2 hours later by intravenous injection of fluorescently labeled antibody to CD45. Three minutes after anti-CD45 injection, mice were euthanized and single-cell suspensions were stained and analyzed by flow cytometry. The OVA peptide did not activate P25TCR-Tg CD4 T cells in either the intravascular (IV+) or the parenchymal (IV−) compartment (Figure 2A), whereas the Ag85B peptide activated parenchymal but not intravascular P25TCR-Tg cells (Figure 2A). In control mice (treated with OVA peptide), in which CD4 T cells are responding to endogenous M tuberculosis antigens, a larger fraction were IFNγ+ in the lung parenchyma than in the vasculature. Ag85B peptide injection increased IFNγ+ cells in both the lung vasculature and parenchyma, although the increase was greater for parenchymal T cells (Figure 2B). These results indicate that intravenously administered peptides activate CD4 T cells in the lung parenchyma, where M tuberculosis-infected cells reside.

Figure 2.

Figure 2.

Open in a new tab

Systemic epitope peptide administration activates lung parenchymal CD4 T cells after loading major histocompatibility complex class II on lung antigen-presenting cells (APC). (A and B) Activation (as determined by intracellular interferon [IFN]γ staining without ex vivo restimulation) of lung CD4 T cells by intravenous administration of Ag85B240-254 compared with OVA323-339 peptide to mice infected 28 days earlier with Mycobacterium tuberculosis. Intravascular and parenchymal lung CD4 T cells were distinguished by intravenous injection of fluorescently conjugated antibody to CD45 immediately before euthanasia; intravascular T cells (IV+) are stained with the intravenous antibody, whereas parenchymal T cells (IV−) are not. (A) Location of adoptively transferred P25TCR-Tg Th1 CD4 T cells that respond to peptide injection as indicated by intracellular staining for IFNγ. (B) Location of endogenous CD4 T cells that responded to peptide injection. (C) Systemic administration of Ag85B240-254 loads peptide on lung antigen-presenting cells in vivo. C57BL/6 mice infected with M tuberculosis H37Rv-EGFP received either Ag85B240-254 or OVA323-339 peptide intravenously on day 28 of infection. One hour later, lung cells were isolated, stained, and subjected to fluorescence-based flow sorting, as previously described [11, 15, 16]. Sorted cells were cultured with Th1-polarized P25TCR-Tg CD4 T cells (10 T cells/APC) without further added antigen for 48 hours, followed by quantitation of T-cell activation by IFNγ enzyme-linked immunosorbent assay. Comparisons between responses to cells from mice that received OVA peptide and those that received Ag85B peptide were by unpaired t test; the P values are given for each comparison. (D) In vivo activation of endogenous CD4 T cells by injection of epitope peptides derived from diverse M tuberculosis antigens. Statistical analysis was by one-way analysis of variance with Dunnett’s posttest, for differences of the effect of each peptide or the peptide pool compared with that of the OVA peptide (*, <.05; **, <.01; ***, <.001). Separate analysis of the peptide pool compared with the individual peptides revealed a nonsignificant difference.

Epitope Peptide Loading of Major Histocompatibility Complex Class II on Myeloid Cells in the Lungs

Because intravenously injected peptides activate parenchymal CD4 T cells, we hypothesized that injected peptides load MHC II on lung cells. To test this hypothesis, we sorted CD11chiCD11bhi dendritic cells (DC) and Ly6G−CD11b+CD11clo recruited macrophages (RM) from lungs of infected mice 1 hour after injection of OVA or Ag85B peptide, and we used them to stimulate P25TCR-Tg CD4 T cells ex vivo. Lung DC from mice that received Ag85B peptide activated P25TCR-Tg CD4 T cells more effectively than did DC from mice that received OVA peptide (Figure 2C). Because the mice were infected with M tuberculosis-EGFP, we also compared the ability of infected and uninfected DC to activate Ag85B-specific CD4 T cells. This revealed that uninfected (EGFP−) DC were more effective at activating Ag85B-specific CD4+ T cells than were infected (EGFP+) DC after injection of the Ag85B peptide (Figure 2C). When RM were compared, infected (EGFP+) RM from mice that received Ag85B peptide activated Ag85B-specific CD4 T cells more effectively than did infected RM from mice that received OVA peptide (Figure 2C). In contrast, uninfected (EGFP−) RM activated Ag85B-specific CD4 T cells to a similar extent, whether they were from mice that received Ag85B peptide or OVA peptide. This is consistent with our finding that uninfected cells acquire Ag85B from infected cells in vivo and present it to CD4 T cells [16], and it suggests that uninfected RM are at their maximum capacity for stimulating CD4 T cells, and injection of Ag85B peptide is unable to further augment their T cell-activating ability. In contrast, infected RM are more impaired in their T cell-stimulating capacity, and systemic administration of Ag85B peptide augments their T cell-stimulating ability.

Epitopes Peptides From Other Mycobacterium tuberculosis Antigens Activate CD4 T Cells In Vivo

A recent study reported that, unlike the Ag85B peptide, administration of an ESAT-6 epitope peptide minimally increased CD4 T-cell production of IFNγ in vivo [7]. Therefore, we determined whether peptides representing epitopes from other M tuberculosis antigens could activate CD4 T cells in infected mice. Injection of Ag85B240-254 peptide or an epitope peptide from HspX increased the frequency of IFNγ+ CD4 T cells, but the changes were not statistically significant after adjusting for multiple comparisons. In contrast, epitope peptides from EsxH (Tb10.4) and Ag85A significantly increased CD4 T-cell activation (Figure 2D). Likewise, a pool of the 4 mycobacterial peptides significantly increased the frequency of IFNγ+ CD4 T cells, but the pool was not more active than any of the peptides alone (Figure 2D).

Mycobacterium tuberculosis Epitope Peptides Possess Modest Antimycobacterial Activity In Vivo

To compare the antimycobacterial activity of diverse peptides in 1 mouse strain, we used CB6F1/J mice, which are heterozygous for MHC alleles (H-2d/b). Mice were infected with M tuberculosis and 28 days postinfection, peptide administration was initiated for a total of 5 doses with 5 days between doses. Six hours after the final dose, bacteria were quantitated in lung homogenates. The antimycobacterial effect of Ag85B peptide 25 was indistinguishable in CB6F1/J mice and in C57BL/6 mice (Figure 3A). Lung CFU in mice treated with each of the M tuberculosis epitope peptides were significantly lower compared with those in mice treated with the control peptide, with the exception of Ag85A104-123 (Figure 3A).

Figure 3.

Figure 3.

Open in a new tab

Limited in vivo antimycobacterial activity of diverse epitope peptides after systemic administration to major histocompatibility complex class biallelic (CB6F1/J; H-2d/b) mice. Mice infected with Mycobacterium tuberculosis H37Rv received intravenous injections of the indicated peptides (50 µg each dose, 5 doses at 5-day intervals) beginning 28 days postinfection. Six hours after the final dose, mice were euthanized and their lungs were homogenized; serial dilutions of the homogenates were plated, and bacterial colonies were counted after 3 weeks of incubation at 37°C. (A) Comparison of antimycobacterial activity of single epitope peptides from 5 M tuberculosis antigens to that of OVA peptide control. Ag85B240-254 was administered to C57BL/6 (B6) and to CB6F1/J (F1) mice to determine the impact of a single I-Ab or 2 I-Ab alleles; no difference was observed between the 2 strains of mice. The other peptides were administered to CB6F1/J mice. Statistical analysis was by one-way analysis of variance (ANOVA) with Dunnett’s posttest, and it reflects differences of the effect of each peptide compared with that of the OVA peptide. (B) Comparison of activity of individual epitope peptides from antigens in which more than 1 epitope has been identified in C57BL/6 (H2b) and/or BALB/c (H2d) mice. Statistical analysis was by one-way ANOVA, comparing the effects of the distinct epitope peptides from a given antigen, to each other. (C) Comparison of activity of single epitope peptides and a pool of the same dose (50 µg each) of the same peptides. Statistical comparison was of the effect of the pool of peptides compared with that of the individual peptides, by one-way ANOVA with Dunnett’s posttest. No statistically significant difference was found. Data in A and B are from the same experiment; data in C are from an independent experiment.

Because individual epitopes derived from a given antigen induce T cells from distinct clones, we compared the in vivo antimycobacterial activity of multiple epitope peptides from HspX, EsxH (Tb10.4), and Ag85A. Comparison of 2 peptides each from HspX and Ag85A, and 3 peptides from EsxH, revealed no significant difference in the antimycobacterial activity of individual epitope peptides from each of these antigens (Figure 3B).

Because distinct epitopes activate different T-cell populations, we reasoned that greater antimycobacterial effects might be obtained by administration of a pool of epitope peptides. Contrary to this prediction, a pool of Ag85B, HspX, EsxH, and Ag85A peptides had no greater antimycobacterial effect than did the individual peptides (Figure 3C).

CD4 T Cells Are Distant From Infected Cells in the Lungs

The modest antimycobacterial effects of epitope peptide administration suggest that factors in addition to antigen availability contribute to the inability of CD4 T cells to eliminate M tuberculosis. Because direct recognition of infected cells by CD4 T cells is essential to control intracellular M tuberculosis [11], we reasoned that one factor that restricts the antimycobacterial efficacy of CD4 T cells is limited colocalization of CD4 T cells and infected cells in the lungs. To assess the spatial relationship between CD4 T cells and infected cells, we examined lung sections from mice infected with M tuberculosis H37Rv-EGFP after staining with antibodies to CD4 T cells and myeloid cells (DC, macrophages, and neutrophils). This revealed that M tuberculosis-infected cells are found as clusters, as expected for bacteria that transfer from cell-to-cell over short ranges [23, 24]. We also found that few M tuberculosis-infected cells have CD4 T cells in direct contact or close proximity (Figure 4A). Indeed, most CD4 T cells are at the periphery of myeloid cell aggregates.

Figure 4.

Figure 4.

Open in a new tab

CD4 T cells are distant from Mycobacterium tuberculosis-infected cells in the lungs. Lung sections from C57BL/6 mice 4 or 8 weeks postinfection with H37Rv-EGFP were first stained for CD11b and CD4 and then nuclear stained with Hoescht dye. (A) Representative image of an M tuberculosis-containing CD11b+ lesion showing lack of proximity of CD4 T cells to M tuberculosis-infected cells. Three arrows highlight bacteria in CD11b+ cells that have no CD4 T cells adjacent. (B) The distance between each bacillus and the nearest CD4 T cell were determined from immunofluorescently labeled lungs from mice 4 or 8 weeks postinfection with M tuberculosis. (C) Representative image of M tuberculosis-infected lung section demonstrating the average distance between M tuberculosis bacilli and CD4 T cells. CD4 T cells are yellow; CD11b+ macrophages, dendritic cells, and neutrophils are blue; Rv-EGFP bacilli are red; and Hoescht+ nuclei are cyan. A total of 23 images from lungs of mice harvested 4 weeks postinfection and 16 images from lungs of mice harvested 8 weeks postinfection, each covering an average area of 696 × 696 μM, were analyzed to determine cell distances.

To quantitate interactions of CD4 T cells and M tuberculosis-infected cells in lung sections, we used a variant of cell distance mapping [25]. We calculated the distances between each bacillus (representing an infected cell) and each CD4 T cell within 100 μM (Figure 4B). The mean distance of the nearest CD4 T cell to each bacillus was 40.93 μM and 39.32 μM at 4 and 8 weeks postinfection, respectively. To determine how these distances relate to CD4 T-cell interactions with infected cells, we estimated the average diameter of CD11b+ cells in the same lung sections by calculating the total area of CD11b+ signal and dividing it by the number of nuclei in the region of CD11b fluorescence. This yielded an average diameter of 7.348 ± 0.7207 μM, smaller than in vitro analyses of macrophage cell diameters. This likely reflects heterogeneity of CD11b+ cell populations, as well as physical constraints within tissues and/or lung granulomas compared with cell culture. These data demonstrate that most CD4 T cells are several cell diameters distant from M tuberculosis-infected cells (Figure 4C). The finding that CD4 T cells are distant from M tuberculosis-infected myeloid cells, together with our previous finding that CD4 T cells must directly recognize infected cells to control intracellular infection [11], suggests that a low frequency of contact may contribute to the inability of CD4 T cells to eliminate M tuberculosis.

DISCUSSION

Mycobacterium tuberculosis uses multiple mechanisms to prevent T cells from eliminating the bacteria [26]. A general mechanism is to limit the activation of T cells at the site of infection; one specific mechanism is to limit the availability of antigen for T-cell activation. We examined the impact of increasing antigen availability on CD4 T-cell activation and control of bacteria in vivo by administration of M tuberculosis epitope peptides. Epitope peptide administration loaded MHC II on lung cells and selectively activated T cells specific for the administered peptide, in the lung parenchyma. Despite activation of lung parenchymal CD4 T cells, epitope peptide administration resulted in only small reductions in mycobacteria. We evaluated the positioning of CD4 T cells and M tuberculosis-infected cells and found infrequent direct contact of CD4 T cells and M tuberculosis-infected cells, suggesting that failure to engage with infected cells may limit the antimycobacterial efficacy of CD4 T cells.

Our finding that epitope peptides are loaded onto antigen-presenting cells to generate peptide:MHC complexes is consistent with the observation that Ag85B peptide 25 injection resulted in arrest of P25TCR-Tg CD4 T cells in Bacillus Calmette-Guérin liver granulomas [6], and it is also consistent with the finding that injection of peptide 25 or ESAT-6 epitope peptide activates CD4 T-cell production of IFNγ in lungs of M tuberculosis-infected mice [7]. Together, the results indicate that activation of antigen-specific CD4 T cells is submaximal during infection with M tuberculosis, and that systemic peptide administration activates CD4 T cells with the properties of a subset that accesses the lung parenchyma and provides antimycobacterial activity in vivo [18]. However, our results indicate that the antimycobacterial activity of these T cells is limited, even after epitope peptide administration.

Although low availability of Ag85B limits T-cell control of M tuberculosis [5], another antigen, ESAT-6, is available to activate CD4 T cells, even late in infection [7]. However, ESAT-6-specific CD4 T cells are driven to exhaustion in humans and in mice, and their ability to proliferate and to produce multiple cytokines is limited [7]. This, together with our finding of limited antimycobacterial activity of epitope peptide administration, indicates that mechanisms beyond antigen availability limit the ability of T cells to eliminate M tuberculosis.

Although our results that CD4 T cells are not closely associated with M tuberculosis-infected cells in the lungs do not prove that this enables persistent infection, our results are consistent with those in rhesus macaques in which CD4 T cells concentrate at the periphery of granulomas, with only a small percentage of the CD4 T cells in the CD11b+ cell-rich central region, where infected cells reside [27]. Our results reveal that the spatial relationship between CD4 T cells and M tuberculosis-infected CD11b+ cells is similar in lungs of rhesus macaques and mice. Considering our finding that CD4 T cells must directly recognize infected cells to control M tuberculosis in vivo [11], we hypothesize that limited recognition and/or contact between CD4 T cells and M tuberculosis-infected cells contributes to suboptimal immunity in tuberculosis.

CONCLUSIONS

Several mechanisms could account for infrequent CD4 T-cell contact with M tuberculosis-infected cells. One is the failure to secrete chemoattractants capable of recruiting CD4 T cells from the lesion periphery to the central core where infected cells reside. Another is that M tuberculosis-infected cells fail to provide a T-cell stop signal [28], causing antigen-specific CD4 T cells to migrate randomly in lesions. An additional mechanism for the failure of CD4 T cells to engage M tuberculosis-infected cells is transfer of antigen from infected to uninfected cells [15, 16, 29]. Because uninfected myeloid cells outnumber infected cells in lungs of M tuberculosis-infected mice [14], uninfected cells that acquire and present bacterial antigens provide abundant targets for recognition by antigen-specific CD4 T cells, thus acting as decoys to minimize T-cell engagement with infected cells. Another possibility is that M tuberculosis-infected cells produce 1 or more signals that repel CD4 T cells, analogous to certain neural guidance molecule ligand-receptor interactions [30]. Yet another hypothetical mechanism is that infected cells selectively kill T lymphocytes, thereby minimizing stable cell contacts and effective T-cell activation. Finally, epithelioid macrophages expressing tight junction proteins [31] may provide a physical barrier that prevents access of T cells to M tuberculosis-infected cells. Further analysis and targeted interventions may reveal the importance of these mechanisms and guide approaches to enhance the antimycobacterial actions of natural and vaccine-induced T-cell responses in tuberculosis.

Notes

Acknowledgments. We thank Colette O’Shaughnessy for contributions during the early stages of this work. We also thank Michael Gregory and Kamilah Ryan of the NYU Cancer Institute Flow Cytometry and Cell Sorting facility (supported by grant P30CA016087) for assistance with live cell sorting. We acknowledge the NIH Tetramer Core Facility (contract HHSN272201300006C) for providing (I-Ab:Ag85B280-294, ESAT-64-17, and EsxG46-61) tetramers.

Finanical support. This work was funded by the National Institutes of Health, National Institute of Allergy and Infectious Diseases (R56 AI105180, R01 AI051242, and T32 AI100853) and a fellowship from the Stony-Wold Herbert Foundation.

Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References

  • 1. Kwan CK, Ernst JD. HIV and tuberculosis: a deadly human syndemic. Clin Microbiol Rev 2011; 24:351–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Lawn SD, Myer L, Edwards D, Bekker LG, Wood R. Short-term and long-term risk of tuberculosis associated with CD4 cell recovery during antiretroviral therapy in South Africa. AIDS 2009; 23:1717–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Urdahl KB. Understanding and overcoming the barriers to T cell-mediated immunity against tuberculosis. Semin Immunol 2014; 26:578–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Urdahl KB, Shafiani S, Ernst JD. Initiation and regulation of T-cell responses in tuberculosis. Mucosal Immunol 2011; 4:288–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Bold TD, Banaei N, Wolf AJ, Ernst JD. Suboptimal activation of antigen-specific CD4+ effector cells enables persistence of M. tuberculosis in vivo. PLoS Pathog 2011; 7:e1002063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Egen JG, Rothfuchs AG, Feng CG, Horwitz MA, Sher A, Germain RN. Intravital imaging reveals limited antigen presentation and T cell effector function in mycobacterial granulomas. Immunity 2011; 34:807–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Moguche AO, Musvosvi M, Penn-Nicholson A et al. Antigen availability shapes T cell differentiation and function during tuberculosis. Cell Host Microbe 2017; 21:695–706.e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Portal-Celhay C, Tufariello JM, Srivastava S et al. Mycobacterium tuberculosis EsxH inhibits ESCRT-dependent CD4+ T-cell activation. Nat Microbiol 2016; 2:16232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Grace PS, Ernst JD. Suboptimal antigen presentation contributes to virulence of Mycobacterium tuberculosis in vivo. J Immunol 2016; 196:357–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Baena A, Porcelli SA. Evasion and subversion of antigen presentation by Mycobacterium tuberculosis. Tissue Antigens 2009; 74:189–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Srivastava S, Ernst JD. Cutting edge: direct recognition of infected cells by CD4 T cells is required for control of intracellular Mycobacterium tuberculosis in vivo. J Immunol 2013; 191:1016–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Tamura T, Ariga H, Kinashi T et al. The role of antigenic peptide in CD4+ T helper phenotype development in a T cell receptor transgenic model. Int Immunol 2004; 16:1691–9. [DOI] [PubMed] [Google Scholar]
  • 13. Wolf AJ, Desvignes L, Linas B et al. 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; 205:105–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Wolf AJ, Linas B, Trevejo-Nuñez GJ et al. Mycobacterium tuberculosis infects dendritic cells with high frequency and impairs their function in vivo. J Immunol 2007; 179:2509–19. [DOI] [PubMed] [Google Scholar]
  • 15. Srivastava S, Ernst JD. Cell-to-cell transfer of M. tuberculosis antigens optimizes CD4 T cell priming. Cell Host Microbe 2014; 15:741–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Srivastava S, Grace PS, Ernst JD. Antigen export reduces antigen presentation and limits T cell control of M. tuberculosis. Cell Host Microbe 2016; 19:44–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Anderson KG, Sung H, Skon CN et al. Cutting edge: intravascular staining redefines lung CD8 T cell responses. J Immunol 2012; 189:2702–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Sakai S, Kauffman KD, Schenkel JM et al. Cutting edge: control of Mycobacterium tuberculosis infection by a subset of lung parenchyma-homing CD4 T cells. J Immunol 2014; 192:2965–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Tagliani E, Shi C, Nancy P, Tay CS, Pamer EG, Erlebacher A. Coordinate regulation of tissue macrophage and dendritic cell population dynamics by CSF-1. J Exp Med 2011; 208:1901–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Moguche AO, Shafiani S, Clemons C et al. ICOS and Bcl6-dependent pathways maintain a CD4 T cell population with memory-like properties during tuberculosis. J Exp Med 2015; 212:715–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Reiley WW, Shafiani S, Wittmer ST et al. Distinct functions of antigen-specific CD4 T cells during murine Mycobacterium tuberculosis infection. Proc Natl Acad Sci U S A 2010; 107:19408–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Sallin MA, Sakai S, Kauffman KD, Young HA, Zhu J, Barber DL. Th1 differentiation drives the accumulation of intravascular, non-protective CD4 T cells during tuberculosis. Cell Rep 2017; 18:3091–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Cambier CJ, Takaki KK, Larson RP et al. Mycobacteria manipulate macrophage recruitment through coordinated use of membrane lipids. Nature 2014; 505:218–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Davis JM, Ramakrishnan L. The role of the granuloma in expansion and dissemination of early tuberculous infection. Cell 2009; 136:37–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Liarski VM, Kaverina N, Chang A et al. Cell distance mapping identifies functional T follicular helper cells in inflamed human renal tissue. Sci Transl Med 2014; 6:230ra46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Goldberg MF, Saini NK, Porcelli SA. Evasion of innate and adaptive immunity by Mycobacterium tuberculosis. Microbiol Spectr 2014 Oct; 2(5). [DOI] [PubMed] [Google Scholar]
  • 27. Kauffman KD, Sallin MA, Sakai S et al. Defective positioning in granulomas but not lung-homing limits CD4 T-cell interactions with Mycobacterium tuberculosis-infected macrophages in rhesus macaques. Mucosal Immunol 2017; doi: 10.1038/mi.2017.60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Dustin ML, Bromley SK, Kan Z, Peterson DA, Unanue ER. Antigen receptor engagement delivers a stop signal to migrating T lymphocytes. Proc Natl Acad Sci U S A 1997; 94:3909–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Samstein M, Schreiber HA, Leiner IM, Susac B, Glickman MS, Pamer EG. Essential yet limited role for CCR2⁺ inflammatory monocytes during Mycobacterium tuberculosis-specific T cell priming. Elife 2013; 2:e01086. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. He Z, Tessier-Lavigne M. Neuropilin is a receptor for the axonal chemorepellent Semaphorin III. Cell 1997; 90:739–51. [DOI] [PubMed] [Google Scholar]
  • 31. Cronan MR, Beerman RW, Rosenberg AF et al. Macrophage epithelial reprogramming underlies mycobacterial granuloma formation and promotes infection. Immunity 2016; 45:861–76. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of Infectious Diseases are provided here courtesy of Oxford University Press

RESOURCES