To explore the involvement of let-7 in emphysema, we scrutinized the genomic locations and transcriptional annotation of let-7 members frequently downregulated in lung T cells isolated from smoker’s lungs as well as mouse models of emphysema. This combined approach showed close linkage and high conservation of two Let-7 clusters encoded from long intergenic non-coding RNA (linc)-like precursors in humans and mice (Figure 1A). To shed light on whether these Let-7 clusters are downregulated in patients with COPD, we analyzed a published (GSE57148) lung RNA-seq dataset obtained from COPD (N=98 and control (N=91) subjects (Kim et al., 2015). Our analysis identified significant downregulation of the Mirlet7ahg and Mirlet7bhg gene cluster transcripts in COPD compared to control subjects (Figure 1B). We carried out quantitative PCR (qPCR) detection of Let-7a, which is encoded by both clusters, in lung tissue samples of smokers with emphysema and non-emphysema controls, detecting significant downregulation of Let-7a in emphysema samples relative to controls (Figure 1C). Because Let-7 has been shown to participate in IL-17⁺ T cell responses (Angelou et al., 2020; Guan et al., 2013; Newcomb et al., 2015), we next sought to determine if the expression pattern of Mirlet7ahg and Mirlet7bhg-derived Let-7 members are impaired in purified CD4⁺ T cells from emphysematous lungs. In support of our original hypothesis, the CD4⁺ T cell expression of Let-7a, Let-7b, Let-7d, and Let-7f were all inversely correlated with more severe emphysema distribution in the lungs as determined by CT scan (Figure 1D).

Next, we elucidated let-7a1/let-7f1/let-7d- and let-7b/let-7c2-clusters expression (herein referred to as let-7afd and let-7bc2 respectively) in murine models of CS- or nCB-induced emphysema respectively (Figure 1E). Paralleling our observations in human COPD and emphysema, mice with CS- or nCB-induced emphysema exhibited reduced expression levels of pri-let7afd and pri-let7bc2 transcripts in the lung and from isolated lung CD4⁺ and CD8⁺ T cells (Figure 1F,G,H). Collectively, our expression results indicate suppression of let-7afd and let-7bc2- clusters in the lung and T cells in human and pre-clinical models of emphysema.

To investigate the in vivo requirement of the let-7bc2-cluster within T cells, we generated conditional ready floxed mice (let-7bc2^flox/flox). We then crossed let7-bc2^flox/flox mice with CD4-Cre mice to generate let-7bc2^flox/flox; CD4-Cre LOF mice (denoted as let-7bc2^LOF mice hereafter) (Figure 2A). This approach allowed us to conditionally delete the let-7bc2 cluster in all T cells derived from the CD4⁺CD8⁺ double-positive (DP) stage (P. Lee et al., 2001; Shi & Petrie, 2012). We confirmed that let-7bc2^LOFmice exhibit robust conditional deletion of the let-7bc2 cluster in DP thymocytes and peripheral CD8⁺ T and CD4⁺ T cells (Figure 2B and data not shown). Our let- 7bc2^LOF adult mice were born at the expected Mendelian frequency and did not show any overt histopathologic or inflammatory changes in lungs histopathology up to 1 year of age in comparison to let-7bc2^f/f control mice (Figure 1-figure supplement 1A,B,C,D and data not shown). Furthermore, quantification of major immune populations and T cell subsets by flow cytometry in let-7bc2^LOF were comparable to control mice under baseline conditions and with moderate aging (Figure 1-figure supplement 1A,B,C,D).

We next exposed let-7bc2^LOF and let-7bc2^f/f control mice to nCB or CS and examined the lungs under the context of experimental emphysema. Histomorphometry measurements of mean linear intercept (MLI) from hematoxylin and eosin (H&E)-stained sections revealed that the enlargement of alveolar spaces sustained from either nCB- or CS-exposure was exaggerated in let-7bc2^LOF mice relative to controls (Figure 2C-E). Chronic inflammation in emphysema is characterized by the recruitment of macrophages and neutrophils to the lung tissue and airways (Peleman et al., 1999; Senior & Anthonisen, 1998). Internally consistent with MLI measurements, let-7bc2^LOF mice treated with nCB showed significantly increased airway infiltration of macrophages and neutrophils in BAL fluid as compared to wild-type control animals (Figure 2F). Concomitant with these findings, expression levels of Mmp9 and Mmp12, which are secreted by macrophages and neutrophils to degrade elastin and mediate alveolar damage, were elevated in airways of let-7bc2^LOF mice exposed to CS versus controls (Figure 2G). Collectively, our data suggests that the let-7bc2-cluster within T cells protects by dampening airway destruction and inflammation because the absence of this cluster worsens the severity of experimental emphysema in mice.

We sought to identify the T cell-intrinsic mechanisms that underlie the exaggerated inflammation observed in emphysematous let7bc2^LOF mice. We focused on the IL-17-mediated T cell response because it promotes neutrophil and macrophage recruitment in the lungs (Beringer et al., 2016; Veldhoen, 2017; Shan et al., 2012). Previously, we established the induction of CD4⁺IL17⁺ (Th17) cells along with CD4⁺IFNg⁺ (Th1) cells in mice with chronic nCB exposure (You et al., 2015), however whether nCB similarly induces CD8⁺IL17a⁺ T cells (Tc17) or cytotoxic T cells (Tc1) had not been studied. The flow cytometric profiling of lung T cells revealed enriched proportions and counts of Tc1 and Tc17 cells in mice with nCB-emphysema and we confirmed the induction of Th17 and Th1 cells (Figure 3A-B control PBS and control nCB). These findings suggests that nCB elicits both the type 17 and type 1 T cell responses, consistent with CS and elastase pre-clinical models of emphysema (Zhang et al., 2019).

We next interrogated the regulatory role of the let-7bc2-cluster in the type 17 and type 1 responses generated from exposure to nCB. Interestingly, let-7bc2^LOF mice showed increased CD8⁺IL17a⁺ Tc17 cells relative to nCB control animals. In contrast, CD8⁺IFN⁺ and GZMA⁺ Tc1 populations remained unperturbed with absence of the let-7bc2 cluster, suggestive of a more refined regulatory role on Tc17 differentiation (Figure 3A-B). There were no significant differences in either Th1 or Th17 cells when comparing nCB-treated let-7bc2^LOF to wild-type controls, indicating the let-7bc2 cluster was dispensable for their generation (Figure 3C). Regulatory T cells form a dynamic axis with Tc17/Th17 cells and act as a counterbalance to lung inflammation in emphysema (Duan et al., 2016; Jin et al., 2014). Therefore, we examined whether Tc17 cell alterations were driven by the let-7bc2 cluster acting on regulatory T cells (Tregs). The let-7bc2^LOF mice showed no significant difference in Tregs subset relative to controls in our model (Figure 3D). Together, our data support the notion that deletion of the let-7bc2-cluster is insufficient to provoke Tc17 cell generation under homeostatic conditions. However, under the context of chronic inflammation in emphysema, the loss of let-7bc2-cluster is intrinsic for the potentiation of T cells towards Tc17 differentiation.

We utilized the TargetScan predictive algorithm to identify putative let-7 microRNA targets that are known to control the IL-17-mediated T cell response (Agarwal et al., 2015). This analysis revealed that the 3’UTR region of Rorc, encoding RORγt, contains an evolutionarily conserved and complementary motif for the let-7 miRNA family (Figure 4A). Thus, we examined if let7bc2- cluster loss in T cells would stimulate and enhance RORγt. Initially, we carried out flow cytometric quantification for RORγt in thymocyte, splenic, and lung T cells of naïve control and let7bc2^LOFmice up to 6-months of age. Our interrogation of RORγt mean fluorescent intensity (MFI) by flow cytometry showed induction of RORγt in single-positive CD8⁺ and CD4⁺ thymocytes, as well as peripheral splenic CD8⁺ and CD4⁺ T cells (Figure 4B). However, RORγt levels appeared unchanged in purified lung CD8⁺ T cells and CD4⁺ T cells of naive let-7bc2^LOF mice, alluding to a compensatory effect in homeostatic lung T cells (Figure 4B). Since we and others have shown that miRNAs are frequently associated with stress-dependent phenotypes, we posited that emphysematous let-7bc2^LOFT cells are poised towards induction of RORγt and production of IL17⁺ subsets after challenge with nCB. Indeed, nCB-emphysematous let-7bc^LOFmice exhibited enhanced RORγt protein levels in both CD8⁺ and CD4⁺ T cells relative to control mice with emphysema (Figure 4C).

Because we had found that the let-7afd-cluster is downregulated in T cells isolated from COPD lungs in human and mice, and that the let-7 family operates with some functional redundancy, we generated mice with conditional deletion of the let7afd-cluster in T cells (let-7afd^f/f; CD4-Cre). The let-7afd^f/f; CD4-Cre (let7afd^LOF) mice aged up to 6-months did not exhibit lung histopathology nor inflammatory changes (data not shown). Of particular interest, ablation of the let-7afd-cluster enhanced levels of RORγt in thymic and peripheral T cells of mice (Figure 4D). Overall, this indicates that independent let-7 clusters restrain RORγt expression levels from thymic development to peripheral T cells under homeostatic conditions. Next, we determined whether loss of let-7afd-cluster in T cells likewise sensitizes mice towards induction of RORγt in nCB-emphysema. Intranasal administration of nCB provoked increased RORγt expression in lung T cells of let-7afd^LOFmice as compared to let-7afd^f/f control mice (Figure 4E), supporting overlapping functionality between the let-7bc2 and let-7afd-clusters in repression of RORγt within T cells.

To confirm that the let-7 family negatively regulates Tc17 cell differentiation, at least in part, cell autonomously in CD8⁺ T cells, we purified naïve CD8⁺ T cells from let-7bc2^LOF and control mice spleens and cultured these cells in vitro in the presence of Tc17 polarizing (TGFβ, IL-6, anti- IFNγ, IL-23, and IL-1β) or Tc1 polarizing (IL-2) conditions (Flores-Santibáñez et al., 2018). Our flow cytometric analysis confirmed the enhanced commitment of let-7-cluster deficient CD8⁺ T cells towards Tc17 cells and IL-17a⁺ production relative to control CD8⁺ T cells (Figure 5A-B). Moreover, enhanced Tc17 cell differentiation mirrored the increased IL-17a detected in the supernatant from in vitro polarized cells as quantified by ELISA (Figure 5C). Parallel assessment of Tc1 differentiation did not detect a difference in CD8⁺IFNγ⁺ cells (Figure 5A and Figure 5D). Altogether, these data recapitulated our in vivo findings that the let-7bc2 cluster negatively regulates Tc17 response but is dispensable in Tc1 cells. Finally, to determine whether Tc17 differentiation is likewise controlled by the let-7afd cluster, we cultured naive CD8⁺ splenocytes from let-7afd^LOF and controls under Tc17 conditions. As we had observed with let-7bc2^LOF, absence of the let-7afd cluster in T cells further enhanced differentiation towards Tc17 cells as quantified by flow cytometry and ELISA (Figure 5E-F).

Next, we focused on Rorc as a potential direct target of let-7, which could mechanistically mediate enhanced Tc17 differentiation in let7bc2^LOF mice. Towards this objective, tested whether let-7bc2^LOFnaïve CD8⁺ T cells show elevated RORγt expression under either Tc0 or Tc17 differentiation conditions. In agreement with enhanced Tc17 differentiation, RORγt expression was differentially and significantly upregulated under both Tc0 and Tc17 differentiation conditions in let-7bc2^LOF cells relative to controls (Figure 5G). To determine whether let-7 directly represses Rorc mRNA levels we cloned the 3’UTR of Rorc into luciferase constructs. These reporter assays with let-7b expressing cells independently confirmed that let-7b represses Rorc (Figure 5H, left). Furthermore, deletion of the putative let-7 binding sequence (Figure 4A) abrogated repression by let-7b (Figure 5H, right), thus confirming Rorc as a functional target of let-7 miRNA. Overall, these in vitro experiments readily recapitulated an upstream regulatory role for let-7 in Tc17 differentiation, mediated in part, via direct suppression of RORγt.

To explore a potential protective role of let-7 miRNA in experimentally induced emphysema, we generated mice which allowed for selective induction of let-7 activity in T cells using the published rtTA-iLet7 mice (Zhu et al., 2011; Belteki et al., 2005). We bred the rtTA-iLet7 mice to CD4-Cre (here in referred to as let-7^GOF) to allow Cre-loxP/doxycycline dependent let-7g overexpression in thymic DP-derived T cells (Figure 6A). Steady-state let-7^GOF and control (rtTA-iLet7) mice were examined for compromised RORγt protein levels within thymocytes and peripheral T cells. Providing further evidence of let-7-dependent regulation of Rorc, protein levels of RORγt were suppressed in CD8⁺ and CD4⁺ T cells of let-7^GOF mice relative to controls (Figure 6B). To determine whether enforced expression of let-7 offered protection from experimental emphysema, let-7^GOF and control mice were treated with nCB and then examined for changes in lung pathology and T cell type 17 responses. The let-7^GOFmice did not exhibit any signs of lung inflammation or pathologic remodeling at baseline (Figure 6C-D and data not shown) Histopathologic analysis revealed a comparable degree of lung alveolar distension via morphometric measurements of MLI in nCB-treated let-7^GOF mice versus controls suggesting that enforced let-7 expression is insufficient to protect the lung from emphysema (Figure 6C-D). On the other hand, evaluation of IL17⁺ response and RORγt levels in emphysematous lung T cells demonstrated that, in contrast to control nCB-treated mice, let-7^GOF mice exhibited dampened lung Tc17 and Th17 cell populations and were resistant to the induction of RORγt after nCB-exposure (Figure 6F-6E). Taken together, our let-7 LOF and GOF models demonstrate the necessity and sufficiency of let-7 miRNA to act as a molecular brake to the type 17 T cell response through the direct regulation of RORγt, further our data suggests that nCB- or CS-mediated suppression of this braking mechanism furthers inflammation and exacerbates emphysema severity (Figure 6G).

Conditional knockout-ready floxed let-7bc2 and let-7afd mice were generated using CRISPR gene editing in an isogenic C57BL/6 genetic background and were sequence verified for rigor. Mice were PCR genotyped from ear samples with primers flanking loxP sites (Supplementary Table). The let-7bc2^flox/flox; CD4-cre and let-7afd^flox/flox; CD4-cre mice were PCR genotyped. The R26-STOP-rtTA; Col1a1-tet0-let-7 (rtTA-iLet7) mice were obtained from JAX Jax Stocks 023912 and 05670 and then bred to CD4-Cre were PCR genotyped with established JAX primers. Control rtTA-iLet7 and the let-7^GOF mice were fed ad libitum with 200mg/kg of doxycycline-containing chow (Bio-Serv S3888) at weaning age. Syngeneic littermates served as controls for all mouse experiments. All mice were bred in the transgenic animal facility at Baylor College of Medicine. All experimental protocols used in this study were approved by the Institutional Animal Care and Use Committee of Baylor College of Medicine animal protocol (AN-7389) and followed the National Research Council Guide for the Care and Use of Laboratory Animals.

Lung tissues were obtained from a total of 28 non-atopic current or former smokers with significant (>20 pack-years, one pack-year equals to smoking one pack of cigarettes per day each year) history of smoking who were recruited into studies from the chest or surgical clinics at Michael E. DeBakey Houston Veterans Affairs Medical Center hospitals (Shan et al., 2009). Human lung single T cells were prepared from surgical resection and lungs in patients as previously described by selection with biotin-labeled antibodies by autoMACs (Miltenyi Biotec) (Yuan et al., 2020; Grumelli et al., 2004). Studies were approved by the Institutional Review Board at Baylor College of Medicine and informed consent was obtained from all patients. Emphysema and non-emphysema control patients were diagnosed from CT scans according to the criteria recommended by the National Institutes of Health–World Health Organization workshop summary (Pauwels et al., 2001).

A publicly available RNA-seq dataset from a Korean cohort GSE57148 was selected for the analysis (Kim et al., 2015). The raw FASTQ files of paired end reads representing the transcriptome of control and cases were retrieved from the GEO database at the National Centre for Biological Information (NCBI) through accession number GSE57148 and analyzed with R package for differential expression.

To promote emphysema, mice were exposed to cigarette smoke using our custom designed whole-body inhalation system (Morales-Mantilla et al., 2020). In total, mice were exposed to four cigarettes (Marlboro 100’s; Philip Morris USA) per day, five days a week, for four months as previously described (Morales-Mantilla et al., 2020, Shan et al., 2012).

Nano-sized particulate carbon black was prepared and administered as previously described (You et al., 2015; Lu et al., 2015). Dried nCB nanoparticles were resuspended in sterile PBS to a concentration of 10 mg/ml. Fifty µl of reconstituted nCB (0.5 mg) were intranasally delivered to deeply anesthetized mice on a schedule of three times a week for four weeks (total delivered dose of 6 mg). Lung histomorphometry and airway inflammation were assessed four weeks after the final nCB challenge. For histomorphometric analysis, mice lungs were fixed with 10% neutral-buffered formalin solution via a tracheal cannula at 25-cm H2O pressure followed by paraffin embedding and tissue sectioning and stained with hematoxylin and eosin. Mean linear intercept (MLI) measurement of mouse lung morphometry were done as previously described (Shan et al., 2014; Morales-Mantilla et al., 2020). Briefly, this was done in a blinded fashion to mice genotypes from ten randomly selected fields of lung parenchyma sections. Paralleled lines were placed on serial lung sections and MLI was calculated by multiplying the length and the number of lines per field, divided by the number of intercepts (Morales-Mantilla et al., 2020).

BALF was collected by instilling and withdrawing 0.8 ml of sterile PBS twice through the trachea. Total and differential cell counts in the BALF were determined with the standard hemocytometer and HEMA3 staining (Biochemical Sciences Inc, Swedesboro, NJ) using 200 μL of BALF for cytospin slide preparation (Morales-Mantilla et al., 2020; Lu et al., 2015).

Mouse lung tissue were cut into 2-mm pieces and digested with collagenase type D (2 mg/ml; Worthington) and deoxyribonuclease (DNase) I (0.04 mg/ml; Roche) for 1 hour in a 37°C incubator. Single-cell suspensions from lung digest, spleen, and thymus were prepared by mincing through 40-μm cell strainers then washing and resuspension in complete RPMI media. Mouse lung and spleen single-cell suspensions were additionally overlaid on Lympholyte M cell separation media (Cedarlane) as indicated in the manufacturer’s protocol to purify lymphocytes. For murine let-7 expression studies, lung single-cell suspensions were labeled with anti-CD4⁺ or anti-CD8⁺ magnetic beads and separated by autoMACS (Miltenyi Biotec), or CD4⁺CD8⁺ double positive cells purified from thymus single-cell suspensions by flow-cytometric sorting on FACS Aria (BD Biosciences).

CD8⁺ naïve T cells were isolated from spleen using Mojosort Mouse CD8 Naïve T cell isolation Kit (Biolegend) and adjusted to a concentration of 1.0x10⁶ cells/mL. Purified cells were activated with plate-bound anti-CD3 (1.5µg/mL) and complete RPMI media containing anti-CD28 (1.5µg/mL) and β-mercaptoethanol (50nM) for Tc0 polarization, or further supplemented with Tc1 [IL-2 (10ng/mL)] or Tc17 [TGFβ (2ng/mL), IL-6 (20ng/mL), anti-IFNγ (10µg/mL), IL-23 (20ng/mL), and IL-1β (5ng/mL)] polarization conditions for 72 hours (Flores-Santibáñez et al., 2018).

Supernatant was collected from in vitro polarized murine CD8⁺ T cells and centrifuged to remove cellular debris. Cytokine levels of IL-17a and IFNγ were quantified from collected supernatant using Mouse IL-17a Uncoated ELISA and Mouse IFN gamma Uncoated ELISA (Invitrogen) Kits, respectively, per the manufacturer’s instructions with colorimetric analysis by the Varioskan LUX microplate reader (ThermoFisher).

Cells used for in vitro or in vivo cytokine analysis were stimulated with PMA (20ng/mL; Sigma Aldrich), Ionomycin (1µg/mL; Sigma Aldrich), and Brefeldin A (2µg/mL; Sigma Aldrich) for 4 hours prior to flow staining (Lu et al., 2015). For intracellular staining, cells were fixed and permeabilized using the Mouse FOXP3 Buffer Set (BD) per the manufacturer’s protocol. The fluorophore-conjugated antibodies used in this study were as follows: Live/Dead Fix Blue (Invitrogen), CD3 PerCPCy5.5 (Biolegend), TCRb PE/Cy7 (Biolegend), CD4 PB (Biolegend), CD4 AF700 (Biolegend), CD8 BV650 (Biolegend), CD25 BV421 (Biolegend), FOXP3 AF488 (Biolegend), ROR gamma T PE (Invitrogen), TCF1 AF647 (Cell Signaling Technologies), TCF1 PE (Biolegend), IFNγ AF647 (Biolegend), IL17a FITC (Biolegend), IL17a PE (ebioscience), PD1 BUV737 (BD), TIM3 AF647 (R&D). Samples were analyzed using BD LSR II flow cytometer (BD Biosciences) and FlowJo software (TreeStar),

RNA was isolated using miRNeasy (Qiagen) or RNeasy Mini Kit (Qiagen) in conjunction with the RNase-Free DNase (Qiagen) according to the manufacturer’s instructions. cDNA of miRNAs and mRNAs were synthesized using TaqMan Advanced miRNA cDNA Synthesis Kit (ThermoFisher) and High-Capacity cDNA Reverse Transcription Kit Real-Time PCR system (Applied Biosystems). 18S and snoRNA-202 were used to normalize mRNA and miRNA expression respectively. Quantitative RT-PCR data were acquired on 7500 Real-Time PCR System or StepOne Real-Time PCR System (Applied Biosystems) with the following TaqMan probes: hsa-let-7a [000377], hsa-let-7b [000378], hsa-let-7d [002283], hsa-let-7f [000382], pri-miR-let7a [Mm03306744_pri], pri-miR-let7d [Mm03306666_pri], pri-miR-let7b [Mm03306756_pri], Mmp9 [Mm00442991], Mmp12 [Mm00500554].

Genomic fragment containing the murine Rorc 3’UTR was cloned into psiCHECK2 luciferase reporter plasmid (Promega). This construct was also used to generate the let-7 ʻseed’ deletion mutant derivative using the QuikChange Multi Site Mutagenesis Kit (catalog 200514-5, Stratagene). 3T3 mouse embryonic fibroblasts (MEFs) were transfected using Oligofectamine (Invitrogen) with 100 ng of psiCheck-2 plasmid containing wild-type or mutant 3’UTR, along with the miRNA control or let-7b duplex (Dharmacon) at a final concentration of 6 nM (Gurha et al., 2012). Reporter activity was detected with the Dual-Luciferase Reporter Assay System (Promega).

Statistical analyses were performed using GraphPad Prism 10.0.1 software. Statistical comparison between groups was performed using the unpaired Student’s t-test, two-way analysis of variance (ANOVA) with Tukey’s or Sidak’s correction, and Mann-Whitney Test when indicated. A P-value less than 0.05 was considered statistically significant; ns indicates not significant. Statistical significance values were set as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Data are presented as means ± SEM. P-value and sample sizes (n) are indicated in the figure legends.