Antigen-specific CD4+ T cells promote monocyte recruitment and differentiation into glycolytic lung macrophages to control Mycobacterium tuberculosis

Although lung myeloid cells provide an intracellular niche for Mycobacterium tuberculosis (Mtb), CD4+ T cells limit Mtb growth in these cells to protect the host. The CD4+ T cell activities including interferon-γ (IFN-γ) production that account for this protection are poorly understood. Using intravenous antibody labeling and lineage-tracing reporter mice, we show that monocyte-derived macrophages (MDMs), rather than phenotypically similar monocytes or dendritic cells, are preferentially infected with Mtb in murine lungs. MDMs were recruited to the lungs by Mtb-specific CD4+ T cells via IFN-γ, which promoted the extravasation of monocyte precursors from the blood. It was possible that CD4+ T cells recruited infectable MDMs because these cells are uniquely poised to receive cognate MHCII-mediated help to control intracellular bacteria. Mice with MHCII deficiency in monocyte-derived cells had normal Mtb-specific CD4+ T cell activation, expansion and differentiation but the CD4+ T cells were unable to attenuate Mtb growth. Using single cell RNA sequencing, we showed that MDMs receiving cognate MHCII-mediated help from CD4+ T cells upregulated glycolytic genes associated with Mtb control. Overall, the results indicate that CD4+ T cells recruit infectable MDMs to the lungs then trigger glycolysis-dependent bacterial control in the MDMs by engaging their MHCII-bound Mtb peptides. Moreover, the results suggest that cognate MHCII-mediated help to promote MDM glycolysis is an essential, IFN-γ-independent effector function of Mtb-specific CD4+ T cells.

Methods
Animals

Mice were housed under specific pathogen-free conditions in accordance with University of Minnesota Institutional Animal Care and Use Committee guidelines. C57BL/6J and B6.SJL-Ptprca Pepcb/BoyJ (CD45.1) mice were bred in-house. The following mouse lines were purchased from Jackson Laboratories and bred in-house: B6.129S2-Tcratm1Mom/J (Tcra-/-), C57BL/6J-Ms4a3em2(cre)Fgnx/J (Ms4a3Cre), B6.Cg-Gt(ROSA)26Sortm9(CAG-tdTomato)Hze/J (tdTomatoLSL), B6.129S7-Ifngtm1Ts/J (IFNg-/-), B6.129S7-Ifngr1tm1Agt/J (IFNgR1-/-), C57BL/6-Ccr2em1(icre/ERT2)Peng/J (Ccr2-CreERT2-GFP), B6.129X1-H2-Ab1b-tm1Koni/J (H2-Ab1fl/fl), B6.129P2-Cd38tm1Lnd/J (Cd38–/–). Ms4a3Cre and tdTomatoLSL mice were bred for one generation to produce mice heterozygous for both alleles. CCR2WT and CCR2ΔMHCII mice were generated by crossing Ccr2-CreERT2-GFP and H2-Ab1fl/fl for several generations to produce litters of H2-Ab1fl/fl mice that were either heterozygous or null for the Ccr2-CreERT2-GFP allele. Littermates were used for experiments. The same approach was used to generate MHCIIdMs4a3 mice and CCR2WT littermate controls. Mice used for experiments were 4-8 weeks old with the exception of bone marrow chimeric mice, which were 4-8 weeks old at the time of irradiation and 14-18 weeks old at the time of infection. CCR2WT and CCR2ΔMHCII mice were fed with chow containing tamoxifen (500 mg/kg diet formulation, Inotiv Teklad) starting on the day of Mtb challenge. Mice were age-matched for experiments and approximately equal proportions of male and female mice were used.

Bacterial strains and plasmids.

The Mycobacterium tuberculosis strain H37Rv was transformed with a chromosomally integrating plasmid expressing a variant of mScarlet fluorescent protein (Bindels, 2017) to generate strain Mtb-mScarlet. This plasmid is selectable by hygromycin resistance, using the mycobacterial optimized promoter (MOP) to drive mScarlet expression and a Giles phage integration module (Morris, 2008).

To make fluorescent H37Rv expressing 2W peptide, an open reading frame encoding H37Rv EsxA (Rv3875) with a C-terminal 4 x glycine linker, followed by 2W peptide (Moon, 2007), followed by a FLAG peptide was inserted into pMN402 (Scholz, 2000). pMN402 contains the promoter region of H37Rv Hsp60 (Rv0440) upstream of the insertion site. The promoter and coding sequence were subcloned into pTT1B, which encodes gentamycin resistance and chromosomally integrates into the Mtb L5 integration locus (Pham, 2007). The resulting plasmid (pTT1B-EsxA-2W) was transformed into Mtb-mScarlet by electroporation and transformants were selected on hygromycin- and gentamycin-containing agar plates.

 Mouse infections.

Mice were infected with a low dose (~100 CFU/lung pair) of Mtb via aerosol exposure. Mtb strains were grown from frozen stocks in complete 7H9 broth (Remel Middlebrook 7H9 broth with 0.2% glycerol, 0.5% BSA, 0.2% dextrose, 0.085% sodium chloride and 0.05% Tween-80). For preparing the bacterial inoculum, Mtb was grown to an optical density at 600 nm (OD600) of 0.4-0.6, washed in PBS-T (PBS containing 0.5% Tween-80) and diluted to an OD600 of 0.005. The inoculum was nebulized for aerosol delivery using the Glas-Col inhalation exposure system. The infectious dose was monitored by plating total lung homogenate the day after aerosol exposure. For CD4+ T cell depletion, α-CD4 or isotype control antibodies were injected intraperitoneally weekly starting on the day of infection.

Processing of infected lungs for flow cytometry and cell sorting.

Mice were euthanized using carbon dioxide inhalation in accordance with University of Minnesota Institutional Animal Care and Use Committee guidelines. Mouse lungs were placed in IMDM complete media (IMDM supplemented with GlutaMAX, pyruvate, non-essential amino acids, 10% fetal bovine serum and 200 µM beta-mercaptoethanol) (Gibco) and homogenized using a GentleMACS tissue dissociator (Miltenyi). Homogenates were strained through a 70 µm filter to generate single cell suspensions.

Cell staining and counting for flow cytometry and cell sorting.

Single cell suspensions were split into multiple samples for T cell and myeloid cell staining. For T cell staining, cells were stained with fluorophore-conjugated antibodies (1:50 dilution), peptide:MHCII tetramers (1:100 dilution), and viability dye (1:500 dilution) for 20 minutes at room temperature. For myeloid cell staining, cells were stained with fluorophore-conjugated antibodies (1:50 dilution) and viability dye (1:500 dilution) for 20 minutes at room temperature. For intracellular staining with iNOS antibody, cells were treated with fixation/permeabilization solution (BD), washed with permeabilization wash buffer (Tonbo), and stained in permeabilization wash buffer for 20 minutes at room temperature. For all samples subject to flow cytometry analysis, cells were fixed overnight in 5% formalin at 4ºC for sterilization. For i.v. labeling experiments, mice were injected retro-orbitally with 2.5 µg of fluorophore-conjugated CD45 antibody 3 minutes prior to euthanasia. To obtain cell counts, AccuCheck Counting Beads (ThermoFisher Scientific) were added to the fixed cell samples immediately before analysis on the flow cytometer.

Monocyte and CD4+ T cell purification and adoptive transfer.

Monocytes purified from donor bone marrow using the Stemcell Mouse Monocyte Isolation Kit were injected retro-orbitally injected into Mtb-infected recipient mice, 2-4 x 106 monocytes per recipient. For purification and adoptive transfer of CD4+ T cells, lymph nodes and spleens of infection-naïve donor mice were pooled, homogenized using a GentleMACS tissue dissociator and strained through a 70 µm filter to generate a single-cell suspension. CD4+ T cells were then purified by negative selection and magnetic separation as follows. Cells in IMDM complete media were incubated with biotinylated antibodies against CD11b, CD11c, CD16/32, B220, CD8a, Ter119, NK1.1 and F4/80 (1:100 dilution) for 15 minutes at room temperature. Stained cells were mixed with streptavidin-conjugated magnetic beads (Stemcell) for 3 minutes, after which a larger volume of complete media was added and the samples were applied to a magnet (Stemcell) for 3 minutes. CD4+ T cells were collected by recovering the unbound fraction. CD4+ T cells purified from donor mice were resuspended in PBS and retro-orbitally injected into infection-naïve recipient mice, approximately 20 million CD4+ T cells per recipient. The CD4+ T cell-recipient mice were rested for 7 days before challenge with Mtb.

Enumeration of bacterial colony-forming units.

Mouse lungs were homogenized as described above. The homogenate was mixed 1:1 with PBS-T without prior straining. Serial dilutions were prepared in PBS-T and plated onto Remel 7H11 agar containing 0.5% glycerol and supplemented with 10% BD BBL Middlebrook OADC Enrichment (containing catalase and oleic acid).

Cytokine quantification.

Mtb-infected mouse lungs were homogenized as described above. A fraction of the homogenate was centrifuged at 10,000 x g for 5 minutes at 4ºC. The supernatant fraction was transferred to a Spin-X 22 µm filter centrifuge tube (Corning) and centrifuged for 30 minutes at 4ºC. The filtrate was collected and stored at -20ºC until assayed. Cytokines were measured in the samples using the LEGENDplex Mouse Inflammation Panel ELISA kit (BioLegend). Concentrations were obtained using standards provided in the kits according to the manufacturer guidelines.

Generation of mixed bone marrow chimeric mice.

Mice were irradiated with 5 x dGly in an X-ray irradiator, rested for 24 hours, and irradiated a second time with 5 x Gly. Bone marrow was collected from donor mice in DMEM media (Gibco) containing 10% fetal bovine serum and counted using a hemocytometer. 2.5 x 105 bone marrow cells from each donor were pooled and transferred into irradiated recipient mice via retro-orbital injection. Mice were rested for approximately 10 weeks prior to infection to allow reconstitution of their hematopoietic systems.

Single cell RNA sequencing.

Single cell suspensions were generated from the lungs of 2 CCR2WT mice (1 male and 1 female) and 2 CCR2ΔMHCII mice (1 male and 1 female) and kept at 4ºC in IMDM complete media. Cells were stained with antibodies against Thy1.2, B220, CD19, Ly-6G, Siglec-F, CD11b, and NK1.1 (1:25 dilution) and viability dye (1:500 dilution) for 30 minutes. Cells staining positively for CD11b and negatively for the remaining markers were sorted using a Sony MA-900 cell sorter. Cells were sorted into IMDM complete media with 40% FBS. Approximately 30,000 cells per sample were recovered from the sorter. Sorted cell samples were processed to obtain RNA in separate wells of the same microfluidics chip using the Chromium Next GEM Single Cell 3’ Gene Expression kit and Chromium X controller (10X Genomics). cDNAs were amplified and libraries barcoded by sample ID were generated according to the manufacturer instructions. Libraries were sequenced by the DNA Services lab at the University of Illinois at Urbana-Champaign. For sequencing, the libraries were pooled, quantitated by qPCR and sequenced on 2 10B lanes with 28 x 150 nucleotide reads on a NovaSeq X Plus with V1.0 sequencing kits (Illumina). The samples each yielded 350-450 million paired reads.

Analysis of RNA sequencing data.

Analysis of single cell RNA sequencing data was performed in R. Using the Seurat package (Hao, 2024), mitochondrial gene features were identified and cells with a mitochondrial feature content of greater than 5% were counted as non-viable and removed from the analysis. Cells with outlying feature counts (under 500 or over 4,000) were also removed. The feature counts for each cell were normalized and scaled using default Seurat parameters as follows. For normalization, the count of each feature was divided by total counts for that cell and multiplied by 10,000. For scaling, features were centered to have a mean of 0 and scaled by the standard deviation of that feature. Scaled data was used to identify variable features, run principal component analysis, and perform differential gene expression analysis..  SingleR (Aran, 2019) was used to identify and remove contaminating lymphocytes, NK cells and PMNs by integrating reference data from ImmGen (Heng, 2008). The clusterProfiler package (Yu, 2012) was used to perform over-representation analysis and GSEA. For GSEA, differentially expressed genes were ranked in decreasing order of log2 fold change. The IFN-γ response module (Fig. 5E) consisted of genes within the mSigDB “interferon gamma response” geneset (Subramanian, 2005) that were enriched in CMs from CCR2WT versus CCR2ΔMHCII mice based on GSEA (genes are listed in Table S3D). For differential gene expression, over-representation and GSEA analyses, statistical significance was determined Wilcoxon rank sum test with Benjamini-Hochberg correction.