We have previously shown that in mice, prior IAV infection leads to elevated pulmonary bacterial burden 16 weeks following subsequent Mtb infection (17). To determine whether prior infection with SCV2 similarly impacts the outcome of Mtb infection we used human Angiotensin Converting Enzyme 2 transgenic (K18-hACE2 Tg) mice, which are susceptible to infection with the ancestral strain of SCV2. K18-hACE2 Tg mice were infected with a sub-lethal dose of the hCoV-19/USA-WA1/2020 (USA-WA1/2020) isolate of SCV2 and 28 days later infected with Mtb ( Figure 1A ). Mtb disease was allowed to develop, and lungs and spleens were collected at 4 weeks ( Figures 1B–D ) or 20 weeks ( Figures 1E–G ) post Mtb infection. SCV2 infection resulted in transient weight loss 5 – 8 days post infection ( Figure 1A ). Lung pathology or bacterial distribution was determined by hematoxylin and eosin (H&E), and acid fast (AF) staining of lung sections 4 weeks post-Mtb infection, however no difference was detected when comparing mice previously infected with SCV2 to those animals that received Mtb alone ( Figure 1B ). Importantly, pulmonary ( Figure 1C ) or splenic ( Figure 1D ) bacterial loads were unchanged in mice previously infected with SCV2 compared to Mtb-only mice. To test whether prior SCV2 infection may affect the control of Mtb at a later timepoint, we assessed lung pathology and bacterial burden at 20 weeks post-Mtb infection. Again, H&E and AF staining did not reveal changes in lung pathology or bacterial localization between mice with prior SCV2 infection compared to mice that were only Mtb infected ( Figure 1E ). Similarly, previous SCV2 infection did not alter pulmonary ( Figure 1F ) or splenic ( Figure 1G ) bacterial loads at this later 20-week timepoint. Taken together, and in contrast to findings with prior IAV infection (17), our data here suggest that prior infection with SCV2 does not lead to increased Mtb-driven disease or impairment of Mtb bacterial replication in mice.

We and others have reported that concurrent or sequential infection with IAV and Mtb resulted in loss of bacterial control (17, 18). To ask whether SCV2 co-infection could equally compromise Mtb replication, K18-hACE2 Tg mice were first infected with Mtb. At a later stage of infection (day 170 post-Mtb) Mtb-infected mice and age-matched controls were then infected with a sub-lethal dose of USA-WA1/2020 SCV2 and monitored for 28 days after which bacterial burdens and lung pathology were assessed ( Figure 2A ). Importantly, SCV2 co-infection did not impact the bacterial burden of Mtb in bronchoalveolar lavage (BAL), lungs, or spleens ( Figure 2B ). Additionally, no differences were seen in lung pathology or bacterial localization as determined by H&E and AF staining ( Figure 2C ) or scoring of affected lung areas ( Figure 2D ).

Next, we asked whether co-infection with SCV2 at 16 weeks post Mtb infection could negatively impact the existing Mtb-specific CD4⁺ or CD8⁺ T cell responses. When we quantified Mtb (ESAT6_4-17)-specific CD4⁺ T cells via MHC-II tetramer straining one month following co-infection with SCV2, the frequency of antigen-specific CD4⁺ T cells was unchanged between mice infected solely with Mtb or those co-infected with SCV2 ( Figure 3A ). There were also no differences in the proportion of lung parenchyma-residing ESAT6_4-17-specific CD4⁺ T cells, as assessed by lack of intravenous CD45 staining (i.v.^neg) (32), nor in the expression of Ki-67 or levels of the transcription factor T-bet on those cells ( Figure 3B ). Likewise, when we examined Mtb-specific CD8⁺ T cell responses the overall abundance ( Figure 3C ) and proportion of parenchymal or KLRG1-expressing cells within Mtb TB10.4_4-11 and Mtb 32c_93-102 MHC-I tetramer positive CD8⁺ T cells were similar between co-infected lungs and lung from mice infected with only Mtb ( Figure 3D ). Together these data suggest that the pre-existing pulmonary Mtb-specific CD4⁺ and CD8⁺ T cell responses are not negatively impacted over the course of 4 weeks following SCV2 co-infection.

We next measured SCV2 antigen-specific T cell responses 4 weeks after SCV2 infection in mice with and without an underlying Mtb infection using SCV2-specific tetramers (33). To measure SCV2 specific CD4⁺ T cells we utilized an ORF3A_266-280 MHC-II I-A^b tetramer and 4 weeks after sub-lethal infection detected approximately 1-2% of effector CD4 T cells that stained positive for the reagent (blue symbols) by flow cytometry, compared to less than 1% in Mtb co-infected mice (purple symbols) and 0.5% non-specific staining background in SCV2 unexposed animals (red symbols) ( Figure 4A ). Thus the overall frequency of ORF3_266-280 specific CD4⁺ T cells was significantly reduced in the lungs of co-infected mice compared to their SCV2-only counterparts. Conversely, proportionately more ORF3_266-280 specific cells were residing in the lung parenchyma (i.v.^neg) and expressed a small but significant increase in T-bet expression ( Figure 4B ). Importantly, both the SCV2 S_539-546-specific and SCV2 N_219-227-specific CD8⁺ T cell responses were significantly reduced in mice with an underlying Mtb infection compared to SCV2 alone ( Figure 4C ). Furthermore, while the overall proportion of i.v.^neg, KLRG1-expressing S_539-546-specific CD8⁺ T cells was the same regardless of Mtb infection status, the frequency of tissue-resident memory (T_RM, CD69⁺) T cells, was significantly reduced within that fraction in co-infected lungs ( Figure 4D ). Thus, ongoing Mtb infection resulted in a significant reduction in the magnitude of the pulmonary SCV2 S_539-546 specific T_RM response.

Considering that antigen burden can directly impact T cell expansion and memory development (34), we asked whether the decrease in SCV2-specific CD8⁺ T_RM frequency 4 weeks after SCV2 infection in Mtb-infected mice was caused by a change in the initial SCV2 viral burden. We suspected that viral loads were reduced in co-infected mice as susceptible K18-hACE2 Tg mice lost 10% of their pre-SCV2 infection body weight 5 – 8 days after SCV2 infection but did not lose any weight if they were also infected with Mtb ( Figure 5A ). To determine whether SCV2 viral titers were reduced in the lungs of co-infected mice, we infected K18-hACE2 Tg mice with USA-WA1/2020 either with or without underlying Mtb infection and collected lungs at 3 days post SCV2 infection, which is early enough to determine viral loads. Indeed, ongoing Mtb infection reduced SCV2 lung viral titers by 1-2 logs at 3 days post-infection as measured by both TCID₅₀ assay on Vero E6 cells ( Figure 5B ) and quantitative PCR (qPCR) to measure the number of copies of the SCV2 E gene in both its actively replicating (sub-genomic, sgRNA) and typical (genomic, gRNA) conformations (35) ( Figure 5C ). Recognizing that the K18-hACE2 model has changes in viral tropism due to the artificial nature of the hACE2 transgene expression (36), we utilized a SCV2 variant of concern (VOC, beta variant, B.1.351), which carries an asparagine to tyrosine substitution at amino acid 501 of the spike protein, allowing binding to murine ACE2 and establishment of transient SCV2 infection in wild type (WT) C57Bl/6 mice (37, 38). Strikingly, we observed a significant reduction in B.1.351 SCV2 viral titers in the lungs of Mtb-infected C57Bl/6 mice as early as 1 day post SCV2 infection, and the magnitude of viral restriction correlated with increasing Mtb-infectious dose, with no replicating virus detectable in the lungs of mice that previously received a high dose of Mtb (1000 - 2000 CFU) ( Figures 5D, E ). Taken together, these results suggest that an underlying pulmonary Mtb infection restricts early viral replication, leading to a reduction in overall viral antigens and a decrease in the magnitude of the T cell and T_RM response.

Because our findings showed restriction of viral replication as early as one day after SCV2 infection in a Mtb dose-dependent manner, we speculated that Mtb-driven innate inflammation, alongside induction of antiviral interferons, may mediate the observed protective effects. Mtb carries several pathogen-associated molecular patterns (PAMPs) that activate pattern recognition receptors (PRRs), including TLR2 and TLR9 (39). TLR activation leads to production of several inflammatory cytokines, including IFN-I. Due to the potent antiviral nature of IFN-I (8), we next examined whether Mtb sensing via TLR2 or TLR9 or a Mtb-driven IFN-I response were required for SCV2 restriction in Mtb-infected mouse lungs. We infected mice deficient in the IFNα receptor 1 (IFNAR1, Ifnar1 ^-/-), TLR2 (Tlr2^-/- ), or TLR9 (Tlr9^-/- ) with Mtb and 1-2 months later with B.1.351 SCV2 ( Figure 6A ). Of note, without underlying Mtb infection (blue symbols), Ifnar1 ^-/- mice displayed a 1.0 – 1.5 log significant increase in viral titers at three days after B.1.351 SCV2 infection compared to WT mice as measured by both TCID₅₀ ( Figure 6A ) and qPCR ( Figure 6B ). SCV2-infected Tlr2^-/- mice had a significant increase in viral titers when measured by TCID₅₀ ( Figure 6A ) but not when assessed by qPCR ( Figure 6B ). Tlr9^-/- mice showed no differences in lung viral titers ( Figures 6A, B ). Importantly and irrespective of these baseline increases in viral titers, we consistently observed a 1.5 – 2.0 log reduction in SCV2 viral loads in lungs of mice with an underlying Mtb infection, regardless of their expression of TLR2, TLR9, or IFNAR1 ( Figures 6A, B ). These results indicate that Mtb-induced restriction of SCV2 is not dependent solely on TLR2, TLR9 or IFN-I signaling and likely is a consequence of multiple innate inflammatory immune alterations during Mtb infection compared to the lungs of immunologically naïve mice.

K18-hACE2 Tg hemizygous transgenic mice (B6.Cg-Tg(K18-ACE2)2Prlmn/J; JAX stock #034860) (36), were purchased from Jackson Laboratories (Bar Harbor, ME). C57Bl/6 mice or C57Bl/6 mice expressing a Foxp3-GFP reporter (C57BL/6-Foxp3^tm1Kuch) (72) were used as wild type C57Bl/6 controls. Foxp3-GFP mice and Ifnar1 KO mice (B6-[KO]IFNa/bR1) (73) were obtained through a supply contract between NIAID and Taconic Farms. Tlr2 KO mice (74) and Tlr9 KO mice (75) were originally generated by the laboratory of Dr. Shizuo Akira (Osaka University, Japan) and were kind gifts of Dr. Alan Sher (NIH/NIAID) and Dr. Giorgio Trinchieri (NIH/NCI) respectively. All mouse strains were confirmed to be on a C57Bl/6 background by genetic background analysis submitted through Transnetyx and performed by Neogen using the MiniMUGA platform. Both male and female mice were used and were 8-16 weeks old at the onset of experiments and mice within experiments were age and sex matched. All animals were bred and maintained in an AAALAC-accredited ABSL2 or ABSL3 facility at the NIH and experiments were performed in compliance with an animal study proposal approved by the NIAID Animal Care and Use Committee.

Aerosol infections of mice with H37Rv-mCherry (50-200 CFU, or as indicated in figure legends) were carried out in a Glas-Col whole-body inhalation exposure system as previously described in detail (76). Briefly, to quantify Mtb CFU, lung or spleen homogenates, BALF or inocula were serial-diluted in PBS + 0.1% Tween-80 and plated on Middlebrook 7H11 agar (Sigma Aldrich) supplemented with oleic acid-albumin-dextrose-catalase (OADC) for 3 weeks at 37°C before colonies were counted.

SARS-CoV-2 hCoV-19/USA-WA1/2020 (Pango lineage A, GISAID reference: EPI_ISL_404895.2) (USA-WA1/2020) and SARS-CoV-2/human/ZAF/KRISP-K005325/2020 beta variant of concern (Pango lineage B.1.351, GISAID reference: EPI_ISL_678615) (B.1.351) were obtained from BEI resources (NIAID, NIH). Viral stocks were generated by infection of Vero cells (CCL-81, American Type Culture Collection) without (USA-WA1/2020) or with (B.1.351) stable expression of TMPRSS2 (77) at a multiplicity of infection of 0.01 for 48hrs. Cell culture media was harvested and centrifuged at 3500 x g, pooled, aliquoted, and stored at -80°C until use. Virus stocks were sequenced using the Illumina platform; USA-WA1/2020 was consistent with the reference sequence MN985325.1 except for H655Y in S, S6L in E, T7I in M, and S194T in N; B.1.351 was consistent with reference sequence MZ376663.1. Mice were anesthetized with isoflurane and infected intranasally with 35µL inoculum containing 1.0x10¹ - 1.0x10³ TCID₅₀ USA-WA1/2020 or 3.5x10⁴ TCID₅₀ B.1.351. Inoculum was quantified by TCID₅₀ assay in Vero E6 cells (CRL-1586; American Type Culture Collection).

2.5x10⁴ Vero E6 cells were seeded in 100μL DMEM + 10% FCS per well of 96-well tissue culture cluster plates, incubated at 37°C + 5% CO₂ for 16-24 hrs and washed twice with 100μL DMEM + 2% FCS before the assay was conducted. After harvesting lungs from mice, the inferior lobe, post-caval lobe and left lung were homogenized in 600μL PBS using 2.7mm glass beads on a Precellys tissue homogenizer (Bertin Instruments) before dilution with PBS to a final volume of 1.7mL. Viral titers were determined by performing 10-fold serial dilutions of homogenates in DMEM + 2% FCS in quadruplicate, then plating 100μL serial-diluted homogenate and 100μL DMEM + 2% FCS on washed Vero E6 cells and incubating at 37°C + 5% CO₂ for 96 hours. TCID₅₀ was measured by removing supernatants and staining wells with crystal violet before scoring for cytopathic effect and calculation using the Reed–Muench method.

For RNA extraction, the superior lobe from each mouse was placed in RNAlater (Invitrogen) and stored at −80°C. RNAlater-stabilized lung lobes were thawed at RT for 20 min, then homogenized in RLT Plus buffer with β-mercaptoethanol (QIAGEN). Total RNA was then isolated from the RLT-homogenized tissue using the RNeasy Plus Mini Kit (QIAGEN), including on-column DNase treatment using the RNase-Free DNase set (QIAGEN) following the manufacturer’s instructions and eluted in 60µL RNAse-free water. SCV2 genome copy quantitation was performed in duplicate from 2.5uL of eluted RNA per reaction using the Taqpath 1-step RT-qPCR Master Mix (Thermo) as described by the manufacturer. The SCV2 E gene in both typical (genomic, gRNA) and actively replicating (sub-genomic, sgRNA) conformations (35) was detected using primers at 500nM as follows: E (genomic) Forward (5’- ACAGGTACGTTAATAGTTAATAGCGT-3’), E (sub-genomic) Forward (5’- CGATCTCTTGTAGATCTGTTCTC-3’), E (genomic and sub-genomic) Reverse (5’- ATATTGCAGCAGTACGCACACA -3’) and the probe for both E genomic and sub-genomic reactions was used at 125nM (5′- (FAM)-ACACTAGCCATCCTTACTGCGCTTCG-(3IABkFQ) -3′). Cycling conditions: Initial: 25°C for 2 min, 50°C for 15 min, and 95°C for 2 min, Cycling: 95°C for 3 sec, 60°C for 30 sec, for 40 cycles. Copy number was calculated based on standard curves generated for each RT-qPCR run, with SCV2 RNA standard of known quantity and eleven 5-fold dilutions run in duplicate (78).

Lungs from infected mice were dissociated using a GentleMACS dissociator (Miltenyi Biotec) in digestion buffer comprised of 0.33mg/mL Liberase TL (Roche), 7U/mL benzonase (Sino Biological), 10µM cytochalasin D (Sigma-Aldrich) and 200µg/mL hyaluronidase (Sigma-Aldrich) followed by 30 - 45 minutes at 37°C. Digested lung was fully dispersed by passage through a 100µm pore size cell strainer and an aliquot was removed for bacterial CFU measurements when needed. Isolated cells were stained with MHC-II tetramers for 40min at 37°C in complete RPMI with 1mM aminoguanidine (Sigma-Aldrich), 100nM dasatinib (Cayman Chemical), 3µg/mL brefeldin A (ThermoFisher) and 2µM monensin (ThermoFisher). Cells were then washed and stained with MHC-I tetramers, surface antibodies and Molecular Probes LIVE/DEAD Fixable Near-IR Dead Cell Stain Kit (ThermoFisher) for 20 min at 4°C before permeabilization and fixation using eBioscience™ Foxp3 Transcription Factor Staining Buffer Set (ThermoFisher) at 4°C overnight. Intracellular staining was performed in eBioscience™ Permeabilization Buffer (ThermoFisher) for 40 minutes at 4°C. Samples were acquired on a FACSymphony (BD Biosciences). FACS data were analyzed using FlowJo10 (Treestar). Antibodies were purchased from BioLegend, BD and ThermoFisher as follows: anti-CD45 (30-F11), anti-CD4 (GK1.5), anti-FoxP3 (FJK-16s), anti-CD8a (53-6.7), anti-CD44 (IM7), anti-Ki-67 (B56), anti-T-bet (4B10), anti-KLRG1 (2F1), anti-CD69 (H1.2F3). Tetramer reagents were obtained from the NIH Tetramer Core Facility as follows: Mtb ESAT6_4-17 MHC-II I-A^b tetramer, Mtb TB10.4_4-11 MHC-I H-2K^b tetramer, Mtb 32c_93-102 MHC-I H2-D^b tetramer, SCV2 ORF3A_266-280 MHC-II I-A^b tetramer, SCV2 S_532-546 MHC-I H-2K^b tetramer, SCV2 N_219-227 MHC-I H-2D^b tetramer.

The middle right lung lobe from each mouse was fixed in 4% paraformaldehyde, transferred to 70% ethanol and paraffin-embedded before sectioning and mounting on glass slides for staining with hematoxylin and eosin (H&E) or the Kinyoun method for visualization of acid-fast (AF) mycobacteria. Stained slides were imaged by light microscopy on an Aperio Versa microscope (Leica Microsystems). Images were processed using QuPath v0.3.2 (Bankhead et al., 2017) and ImageJ v1.53t (NIH) for quantification and visualization as previously described (79).

Statistical analyses were performed using GraphPad Prism v9.0 for Mac OS X (GraphPad Software). Each figure legend lists all the statistical details of experiments, including the statistical tests used. Data are expressed as mean ± SD. Significant differences are indicated by the p value in each figure.