HIV disrupts the lung molecular clock, leading to lung inflammation and features of emphysema

Globally, ~39 million people are estimated to be living with human immunodeficiency virus (HIV)1. While access to combination antiretroviral therapy (cART) has increased, a cure remains elusive, as cART alone is incapable of eradicating HIV reservoirs2. In the aging, HIV-positive population, comorbid diseases are essential determinants of morbidity and mortality3. People living with HIV (PLWH) receiving effective cART tend to die from non-acquired immune deficiency syndrome (AIDS)-related comorbidities nearly a decade earlier than those without HIV4. Chronic obstructive pulmonary disease (COPD) is a major comorbidity in PLWH, with HIV infection serving as an independent risk factor. The progression and severity of COPD are further worsened by cigarette smoking5.

COPD exacerbations demonstrate a diurnal rhythm, with more frequent exacerbations occurring during the night or early morning hours, likely due to circadian dysregulation affecting mucus production and lung inflammation6,7,8. The circadian clock constitutes a nearly 24 h oscillation in gene expression, leading to changes in physiological processes, metabolism, and behavior, and thereby orchestrating the rhythmic synchronization of physiological processes9,10,11. In mammals, the circadian timing system consists of a central pacemaker clock located in the suprachiasmatic nucleus (SCN) of the brain, along with additional peripheral clocks found in most tissues throughout the body12. Although light/dark cycles are the primary zeitgebers (timing cues) for the SCN pacemaker that subsequently synchronizes peripheral clocks, feeding patterns and environmental factors can also act as potent zeitgebers, influencing the tuning or disruption of peripheral clocks13.

In the lung, the circadian rhythm regulates daily fluctuation in airway diameter, airway resistance, respiratory symptoms, mucus production, and inflammatory responses, which in turn contribute to the diurnal pattern of exacerbation frequency and occurrence14,15.

The molecular clock operates through a negative feedback loop involving various transcription factors known as clock genes. In mammals, the CLOCK:BMAL1 heterodimer promotes the expression of period (PER1-3) and cryptochrome (CRY1-2) genes. PER and CRY form phosphorylated heterodimers that translocate back to the nucleus, where they inhibit the activity of the BMAL1:CLOCK complex16,17.

Post-translational modifications including as acetylation and phosphorylation, play critical roles in regulating the activity and stability of molecular clock proteins. In the lung, clock dysfunction can lead to alterations in pulmonary physiology and contribute to disease pathologies, particularly in response to environmental stressors such as cigarette smoking14. Sirtuin 1 (SIRT1) is the primary circadian deacetylase linking core clock functions with inflammatory responses in the lung. It deacetylates BMAL1 and PER2, modulating their activity and, by extension, sustaining the circadian expression and activity of these clock proteins. The suppression of SIRT1 in airway epithelia of smokers and COPD patients is associated with a dysregulated molecular clock and consequent lung inflammation and injury10,18,19. Recent studies, including ours, show that clock dysfunction in airway epithelial cells plays an important role in pulmonary physiology and pathology, particularly in response to proinflammatory mediators such as cigarette smoking14,18.

MicroRNAs (miRNAs) play an important role in post-transcriptional regulation of genes and are often dysregulated in disease20. HIV transactivator of transcription (TAT) is a viral protein secreted by infected cells that enters bystander cells and alter gene expression independent of viral replication. In our study, we employed both whole HIV and purified Tat protein to delineate TAT-specific effect. We previously reported that HIV TAT alters the bronchial epithelial microRNAome21. Here, we identify novel miR-126-3p-mediated SIRT1 regulation as being,primarily responsible for HIV-mediated dysregulation of the lung molecular clock. These observations were validated in primary normal human bronchial epithelial cells (NHBE), SP-C TAT transgenic mice (lung-specific TAT expression), and HIV-positive human lungs. We further demonstrate that this dysregulation has downstream effects on the lung, including lung inflammation. ScRNAseq experiments on lungs from young adult SP-C TAT transgenic mice demonstrated that HIV TAT expression, resulting in circadian disruption and inflammation, becomes evident in multiple cell types at very early stages of life. The significance of our findings will be discussed.

Circadian genes are suppressed in lungs from HIV-positive donors

Total RNA was quantified for expression of major circadian genes. Quantitative reverse transcription-PCR(qRT-PCR) analysis revealed a significant reduction in the expression of major circadian clock genes in HIV-positive lung tissues compared to HIV-negative controls. Specifically, there was a statistically significant downregulation of PER1, PER2, REV-ERB α, CRY2, BMAL1, CLOCK, and RORA (Fig. 1A–H). Interestingly, the expression of CRY1 also trended lower in HIV-positive paired donor lungs, although the suppression was not significant (Fig. 1G, H). While expression of circadian genes in non-HIV control lungs displayed greater inter-individual variability, consistent with known sensitivity of peripheral circadian rhythms to lifestyle and physiological factors, HIV-positive lung samples exhibited a more uniform suppression of circadian genes reflecting the dominant influence of viral proteins such as Tat on circadian regulation, which may override inter-individual variability.

Fig. 1: Lung tissues from HIV-positive donors demonstrate suppression of circadian genes.

A–H mRNA expression levels (fold change) of key clock genes in human lung tissue samples from HIV-negative control and HIV-positive donors. A significant reduction in expression of PER1, PER2, REV-ERB α, CRY2, BMAL1, CLOCK, and RORA was observed in the HIV-infected group compared to controls, as determined by qRT-PCR analysis. Statistical significance was determined using Student’s t-test (*p < 0.05, **p < 0.01, ***p < 0.001; ns=non-significant). Data are shown as mean of n = 5 ± SEM.

Full size image

To determine if our HIV positive samples demonstrate Tat expression in the lungs we performed an Immunohistochemistry (IHC) analysis using a HIV Tat-specific antibody. Tat immunoreactivity was localized to the alveolar and airway epithelium and was accompanies by infiltration of inflammatory cells only in lung sections from HIV positive donors. No detectable Tat staining was observed in HIV-negative lung tissue (Supplementary Fig. 1). HIV-positive lungs show approximately sixfold increase in staining intensity compared to HIV-negative controls.

HIV TAT dysregulates the expression of BMAL1 and PER2 in NHBE ALI cultures

Previous studies have shown that levels of HIV TAT in the cerebrospinal fluid of PLWH negatively correlate with circadian rhythms, affecting sleep as well as blood pressure22, and that TAT expression in the brain alters circadian regularity and activity23. Our findings, along with previous reports, have demonstrated that the bronchial epithelium expresses canonical HIV entry receptors and is permissive to HIV infection24,25. HIV infection of NHBE ALI cultures demonstrated suppression of BMAL1 and PER2 while also showing a weak amplitude of circadian oscillations compared to the control cultures (Fig. 2A, B).

Fig. 2: HIV infection disrupts circadian rhythm, and HIV TAT suppresses expression of core clock genes in primary bronchial epithelial cells.

A, B NHBE ALI culture were infected with 5 ng p24, equivalent to HIV BaL strain. The molecular clock was synchronized 24 h post-infection using forskolin (10 µM/30 min). Cells were washed to remove forskolin, and experiments were terminated at designated time points following synchronization. Total RNA was analyzed for BMAL1 and PER2 mRNA. HIV suppressed the expression and circadian amplitude of BMAL1 and PER2; n = 4. Statistical significance was determined using Student’s t-test (***p < 0.001, ****p < 0.0001). C–J Primary NHBE ALI cultures were treated with TAT or heat-inactivated TAT as control. Total RNA was extracted 48 h post-TAT treatment and quantitated by qRT-PCR for core clock genes (BMAL1, CLOCK, PER2, CRY2, REV-ERBα, RORA, PER1, and CRY1). HIV TAT demonstrated a significant decrease in all core clock genes compared to controls. Data are shown as mean ± SEM from three different lungs. Statistical significance was determined using Student’s t-test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns=non-significant).

Full size image

To determine any specific effects of TAT in disrupting the circadian rhythms of airway epithelia, we treated NHBE ALI cells with recombinant HIV TAT and quantified the expression of clock genes. All the core clock genes (PER1, PER2, BMAL1, CRY1, CRY2, REV-ERBα, and RORΑ) showed significant suppression following TAT treatment compared to the control (Fig. 2C–I). Although the suppression of CLOCK trended lower with TAT treatment, it was not significant (Fig. 2J). To understand the temporal effects of HIV TAT on the circadian expression pattern of BMAL1 and PER2 at the protein level, the circadian clock was synchronized following TAT treatment. The expression of the BMAL1 and PER2 proteins in the TAT-treated group demonstrated a significant decrease compared to the heat-inactivated TAT controls (Fig. 3A, B). Both BMAL1 and PER2 proteins showed higher expression at the 12 h time point and lower expression at the 24 h time point. Lower protein expression was observed with TAT treatment at 0, 12, and 24 h for PER2, while maximum downregulation of BMAL1 was observed at 24 h.

Fig. 3: HIV TAT decreases levels and oscillation amplitude for SIRT1 deacetylase targets of core clock genes BMAL1 and PER2.

Primary bronchial epithelial cells were treated with HIV TAT or heat-inactivated TAT as a control. The cellular circadian clock was synchronized 24 h post-treatment with 10 µM forskolin for 30 min. Experiments were terminated at 0, 12, and 24 h post-synchronization, and protein levels of PER2 (A) and BMAL1 (B) were analyzed using western blot analysis. HIV TAT decreased levels and oscillation amplitude of BMAL1 and PER2 proteins in the TAT-treated group compared to controls. Data are shown as mean ± SEM from three different lungs and protein expression levels were normalized to GAPDH.

Full size image

HIV TAT suppresses SIRT1 in NHBE ALI cells

SIRT1 is a master clock deacetylase26. We have previously demonstrated that disruption of SIRT1-mediated BMAL1 regulation in the epithelium plays an important role in cigarette smoking-induced lung inflammation18. SIRT1 is a highly conserved class III histone deacetylase that is involved in deacetylation of several proteins27. Kwon H-S et al. (2009) showed that TAT promotes hyperactivation of T-cells by inhibiting the SIRT1 activity28. While this interaction was at the protein level, we wanted to explore whether other mechanisms exist by which HIV dysregulates SIRT1. We previously reported that HIV TAT dysregulates the bronchial epithelial microRNAome29. We explored whether HIV TAT and, by extension, HIV infection, suppresses SIRT1 biogenesis in bronchial epithelial cells and if this is due to miRNA-mediated silencing. We first investigated if SIRT1 levels are suppressed in PLWH. We quantitated SIRT1 mRNA levels in control lungs and lungs from HIV-positive donors. Figure 4A shows that SIRT1 is suppressed in lungs of HIV-positive donors compared to HIV-negativecontrols.

Fig. 4: HIV TAT suppresses SIRT1 expression and upregulates miR-126-3p in primary bronchial epithelial cells and HIV-positive donor lungs.

A mRNA expression levels (fold change) of SIRT1 in lung tissue samples from non-smoker HIV-negative control donors and HIV-positive donors. Data are shown as mean of n = 5 ± SEM. B Primary NHBE cells were infected with 5 ng p24, equivalent of HIV BaL (R5-tropic strain). Experiments were terminated after 48 h, and total RNA was analyzed. Fold change expression of SIRT1 in R5-infected NHBE cells was compared to control uninfected NHBE cells. Data are shown as mean ± SEM from three different lungs. C Primary bronchial epithelial cells were treated with HIV TAT or heat-inactivated TAT as control. A separate set was treated with ATA, which rescues HIV TAT suppression of SIRT1 mRNA. Data are shown as mean ± SEM from three different lungs. D NHBE ALI cultures were treated with HIV TAT or control. HIV TAT suppressed SIRT1 protein levels (western blot analysis); Data are shown as mean ± SEM from five different lungs; *significant. E An in silico analysis using miRTARbase and prediction by Miranda confirmed the presence of the miR-126-3p site on the 3’UTR of SIRT1 mRNA, suggesting a direct regulatory relationship between miR-126-3p and SIRT1. F NHBE ALI cultures were treated with HIV TAT or heat-inactivated TAT as control. qRT-PCR analysis revealed a significant fold change in miR-126-3p expression, demonstrating its upregulation compared to control. Data are shown as mean of n = 3 ± SEM. G miR-126-3p expression was analyzed in lung tissues from HIV patients and compared with uninfected lungs. The qRT-PCR results indicated a significant upregulation of miR-126-3p in the lungs of HIV patients. Data are shown as mean of n = 3 ± SEM. H BEAS-2B cells were transfected with 40 nM of miR-126-3p mimic using electroporation. qRT-PCR analysis revealed a significant decrease in SIRT1 mRNA levels. Data are shown as mean of n = 3 ± SEM. I Western blot analysis confirmed the downregulation of SIRT1 protein in BEAS-2B cells upon transfection with miR-126-3p mimic; Data are shown as mean of n = 3 ± SEM; *significant.

Full size image

To further validate these findings, we infected NHBE ALI cultures with HIV and quantitated SIRT1 RNA levels. Figure 4B demonstrates that SIRT1 is downregulated in the HIV-infected NHBE ALI cultures. Our data show that HIV infection suppresses SIRT1 both in vivo in human lung tissue and in vitro in NHBE ALI cultures. NHBE ALI cells were treated with HIV TAT (or control). Another experimental group was pre-treated with aurintricarboxylic acid (ATA), a small molecule inhibitor of the miRNA processing enzyme Drosha30. Total RNA was quantitated for SIRT1 expression 48 h post TAT treatment. The qRT-PCR data confirmed the downregulation of SIRT1 mRNA and its reversal by ATA, suggesting that TAT suppresses SIRT1 by miRNA-mediated regulation (Fig. 4C). Similarly, NHBE ALI cultures treated with HIV TAT demonstrated decreased levels of SIRT1 protein similar to mRNA suppression (Fig. 4D).

HIV TAT upregulates miR-126-3p, suppresses SIRT1, and increases acetylated BMAL1

We previously showed that HIV TAT treatment leads to statistically significant changes in the expression of 83 different microRNAs in bronchial epithelial cells21. We used miRTARbase and miRanda to identify potential miRNA-SIRT1 pairs31. Using miRanda, the miRNA-target prediction confirmed the presence of a complementary sequence (including the seed sequence) at position 922–943 for the miR-126-3p binding site on the 3’UTR of the SIRT1 mRNA (Fig. 4E). This sequence is highly conserved among several vertebrates, suggesting a potential novel miRNA-mRNA interaction for microRNA-mediated SIRT1 regulation. We revisited our earlier miRNA array data and noticed that HIV TAT also upregulated miR-126-3p (log2 fold change of 4.81; p-value of 0.059, narrowly missing statistical significance; supplementary data in ref. 21). We re-examined this finding to determine if miR-126-3p plays a role in HIV TAT-mediated SIRT1 suppression and observed that miR-126-3p levels were significantly increased in TAT-treated NHBE cells (Fig. 4F). Lung tissues from HIV-positive donors also demonstrated significant upregulation of miR-126-3p compared to the controls (Fig. 4G). Finally, we validated the miR-126-3p-SIRT1 regulatory axis by transfecting miR-126-3p miRNA mimics (or control) in BEAS-2B airway epithelial cell lines. Our data show that miR-126-3p transfection suppresses SIRT1, thereby confirming the regulation of SIRT1 by miR-126-3p (Fig. 4H, I). Together, these data identify miR-126-3p as a novel microRNA candidate for HIV TAT-mediated SIRT1 suppression.

Nakahata et al. showed that SIRT1 regulated the activity of the CLOCK-BMAL1 heterodimer by deacetylating BMAL132. We have previously shown that dysregulation of the SIRT1/BMAL1 axis in airway epithelial cells causes lung inflammation in mice exposed to cigarette smoke18. To determine if SIRT1 suppression translates to changes in the acetylation status of BAML1, its deacetylase target, we transfected NHBE cells with siSIRT1. SIRT1 silencing was confirmed by qRT-PCR and western blot analyzes of SIRT1 levels (Supplementary Fig. 2A, B). SIRT1 silencing led to a significant increase in levels of Ac-BMAL at Lys538 compared to the control (Supplementary Fig. 2C). Ac-BMAL1 further manifests as upregulation of the IL-6 gene, as previously reported33,34 (Supplementary Fig. 2D).

HIV TAT disrupts circadian oscillation by upregulating miR-126-3p

To investigate the role of TAT and miR-126-3p on PER2 expression and circadian oscillation in bronchial epithelial cells, we transduced NHBE cells with the lentiviral promoter reporter vector, PER2-Luc, in which the PER promoter drives the firefly luciferase reporter. Undifferentiated bronchial epithelial cells were used since we have previously shown efficient transduction of undifferentiated cells with lentiviral vectors35, but poor transduction of differentiated bronchial epithelial cells (not shown). Firefly luciferase is used extensively across various organisms and cell types to study circadian rhythm36,37,38. To determine the role of miR-126-3p in dysregulating circadian rhythm via SIRT1 suppression, we transfected PER2-transduced NHBE cells with miR-126-3p. Our data show circadian expression of luciferase in respective controls with maximal PER2 promoter activity recorded at 12-h and 36-h time intervals. Treatment with HIV TAT (Supplementary Fig. 3A) or transfection with miR-126-3p mimic (Supplementary Fig. 3B) significantly depressed the amplitude and levels of luciferase expression driven by PER2 promoter expression compared to the controls. SIRT1 levels were quantitated for the peak amplitude time point (36 h) and showed that HIV TAT and miR-126-3p suppress SIRT1. These data demonstrate that HIV TAT induces miR-126-3p to suppress SIRT1 with consequent downstream effects on lung circadian gene expression. We have used undifferentiated as they show increased transfection efficiency compared to fully differentiated NHBE ALI culture. We do not anticipate that the use of differentiated cells would alter the overall results or conclusions, as the mechanism of RNA interference is highly conserved across cell types and species.

TAT-mediated disruption of circadian rhythm leads to increased secretion of proinflammatory cytokines

We have demonstrated that disruption of the SIRT1/BMAL1 axis in the airway can lead to increased inflammation18. NHBE ALI cultures were treated with TAT or control for 48 h as described in the methods. Similarly, another set of cells was transfected with miR-126-3p or control miR-NC. TAT treatment significantly increased proinflammatory cytokines in NHBE ALI cultures compared to the control (Supplementary Fig. 4A). Specifically, we observed increased levels of TNF-α, IL-6, IL-8, GM-CSF, bFGF, IFN-γ, IL-4, and IL-17 in the TAT-treated cells compared to controls. Transfection with miR-126-3p mimics upregulated only MCP-1 and IFN-γ (Supplementary Fig. 4B; both cytokines were also upregulated by TAT) compared to multiple cytokines upregulated by HIV TAT. Separately, the fold change in CXCL5 mRNA expression was higher in TAT-treated NHBE cells compared to control cells as determined by qRT-PCR (Supplementary Fig. 4C). These cytokines play a significant role in lung inflammation, specifically in patients with COPD39,40,41,42,43,44,45.

SIRT1 knockout disrupts circadian protein expression and enhances HIV-1-induced inflammation

To investigate the role of SIRT1 in regulating circadian clock proteins and inflammation, we generated SIRT1 knockout NHBE cultures using CRISPR/Cas9 and confirmed the knockout via western blot analysis. SIRT1 protein levels were significantly reduced in the knocked-out cells compared to the control group (Fig. 5A); however, a low level of residual SIRT1 expression was still detected, indicating incomplete knockout. Next, we examined the SIRT1 depletion on core circadian components. Western blot analysis revealed a significant reduction in PER2 (Fig. 5B) and BMAL1 (Fig. 5C) protein levels upon SIRT1 knockout, indicating that SIRT1 is essential for maintaining circadian rhythm stability in NHBE ALI cultures and directly regulates the core clock genes PER2 and BMAL1.

Fig. 5: The effect of SIRT1 knockout and SRT1720 treatment on circadian gene expression and cytokine secretion in NHBE ALI.

A Western blot analysis confirming SIRT1 knockout in NHBE ALI using CRISPR/Cas9. A significant reduction in SIRT1 protein expression is observed compared to the control group. B Western blot analysis of PER2 protein levels following SIRT1 knockout. PER2 expression is significantly reduced, indicating that SIRT1 positively regulates PER2. C Western blot analysis of BMAL1 protein levels upon SIRT1 knockout. Data are shown as mean ± SEM from three different lungs. D Luminex data showing the secretion of proinflammatory cytokines in NHBE ALI under different conditions. HIV-1 infection significantly increased cytokine levels compared to the control group. Pre-treatment with 1 µM SRT1720 (R5+SRT1720) led to a marked reduction in cytokine levels compared to the R5-infected group.

Full size image

Additionally, we assessed the inflammatory response in NHBE cultures in different conditions using Luminex analysis. HIV-1 (R5) infection led to a significant increase in proinflammatory cytokine secretion compared to the control group (Fig. 5D), demonstrating a virus-induced inflammatory response. However, pre-treatment with 1 μM SRT1720 (R5+SRT1720), a potent SIRT1 activator, resulted in a marked reduction in cytokine levels compared to the R5-infected group. Analysis of proinflammatory cytokines from the culture supernatant demonstrated that HIV infection increased expression of IL-6, IL-8, G-CSF, GM-CSF, MCP-1, and IP-10. Treatment with SRT1720 reversed these increases in proinflammatory cytokines. In addition, we also observed a marked reduction in the baseline levels of several key proinflammatory cytokines, including IL-5, RANTES, VEGF, and IL-4 (Fig. 5B).

Circadian gene suppression by HIV TAT leads to lung inflammation and features of emphysema in SP-C TAT mice

SP-C mice were first generated and described by Cota-Gomez et al.46, with TAT (86-amino acid isoform) expressed under the control of surfactant protein-C promoter (SP-C), thereby restricting TAT expression to the lung and, more specifically, the small airways. HIV TAT contains a protein transduction domain that allows it to be secreted and taken up by other cells47,48,49,50, and can have pleiotropic effects on other lung cells as well. PLWH demonstrates the onset of COPD symptoms as early as age 38-4751,52. Lungs from 7-month-old SP-C TAT mice (~35 years human equivalent53) were analyzed for expression of major circadian genes and SIRT1. SP-C TAT mice demonstrated a significant downregulation of core circadian genes SIRT1, BMAL1, PER1, PER2, CRY1, CRY2, REV-ERB α, CLOCK, and RORA normalized with GAPDH (Fig. 6A–G). These results are similar to those observed by us in lungs from HIV-positive lung donors (Fig. 1A–H), confirming the role of HIV TAT in circadian disruption in the lung. Western blot analyzes demonstrated suppression of SIRT1 protein levels (Fig. 7A) and its circadian deacetylase targets BMAL1 and PER2 (Fig. 7B, C) in lungs from SP-C TAT mice compared to WT littermate controls. IHC analysis of SPC-TAT mice was carried out to determine the impact of HIV-related stress on tissue architecture and molecular clock protein expression under in vivo conditions. Lungs from control and SP-C TAT-expressing mice were examined using specific antibodies against molecular clock proteins. Histological evaluation of airway/alveolar and parenchymal lung sections revealed significant downregulation of molecular clock proteins (PER2, SIRT1, and REV-ERBα), consistent with mRNA expression data (Fig. 8A–C). Surprisingly, BMAL1 does not show any significant downregulation in IHC staining (data not shown). Protein expression was localized primarily in the epithelial and basement membrane regions, suggesting that these clock proteins regulate essential cellular functions in these areas. Moreover, we also observed abnormal enlargement of the alveolar airspaces and epithelial hyperplasia in the airway, with the cells surrounding the airway regions exhibiting prominent thickening. Lungs from SP-C TAT mice demonstrated increased ROS damage and airspace enlargement (evident from the mean linear intercept) compared to WT littermate controls, characteristic of early stages of emphysema (Fig. 8D, E). These data demonstrate that HIV TAT-mediated circadian dysregulation leads to lung inflammation and features of emphysema.

Fig. 6: Lungs from SP-C TAT mice demonstrate suppression of circadian genes.

A–I HIV TAT affected major clock genes and SIRT1 gene expression in SPC-TAT mouse lungs. Total RNA from whole lungs of 6-month-old SP-C TAT mice, with WT littermates as control, were analyzed for major clock gene expression. Expression of core clock genes, including BMAL1, CRY1, CRY2, PER1, PER2, CLOCK, RORA, and REB-ERBα was examined by qRT-PCR. Lungs of SPC-TAT mice showed suppression of all the clock genes, including SIRT1. P < 0.001 for PER1; p < 0.001 for CRY1 and REV-ERBα; p < 0.01 for BMAL1, PER2, and CRY2; p < 0.05 for SIRT1; ns for CLOCK and RORA. Data are shown as mean of n = 7 ± SEM.

Full size image

Fig. 7: Western blot analysis of lung tissue from SP-C TAT and wild-type control mice demonstrating the suppression of key circadian clock proteins.

A Western blot analysis of lung tissue from 6-month-old SP-C TAT mice and wild-type control mice showed a significant reduction in SIRT1 protein levels in SP-C TAT mice compared to controls (Data are shown as mean of n = 4 ± SEM; 2 males and 2 females). B Western blot analysis confirmed the decreased expression of PER2 protein in the lung tissue of SP-C TAT mice compared to wild-type controls (Data are shown as mean of n = 3 ± SEM; 2 males and 1 female). C Reduced BMAL1 protein level observed in the lung tissue of SP-C TAT mice compared to wild-type control (Data are shown as mean of n = 3 ± SEM; 2 males and 1 female).

Full size image

Fig. 8: SP-C TAT mice exhibit dysregulation of molecular clock proteins, oxidative stress, and lung tissue remodeling.

Immunohistochemical (IHC) analysis was performed on lung sections from control and SPC-TAT mice using specific antibodies against key molecular clock proteins. REV-ERBα (A), SIRT1 (B), and PER2 (C) expression was assessed in airway, alveolar, and parenchymal lung regions. Protein expression was primarily localized in the epithelial and basement membrane regions, indicating their role in cellular homeostasis. SPC-TAT mouse lungs exhibited significant downregulation of these clock proteins, accompanied by structural abnormalities, including alveolar airspace enlargement, epithelial hyperplasia in airway regions, and airway wall thickening. D Paired lung sections from SP-C TAT transgenic mice or WT littermates were deparaffinized and rehydrated, followed by antigen retrieval. Lung sections were stained with anti-4-HNE antibody (Abcam, cat. no. ab48506). Slides were developed with DAB Quanto Chromogen and substrate (Thermo Fisher Scientific, cat. no. TA-125-QHDX) and then counterstained with hematoxylin. SP-C TAT mice showed increased 4-HNE levels, indicative of ROS damage by carbonylation. Staining intensity was quantitated using ImageJ. The relative IHC score assessment based on positive staining intensity was performed in a semi-quantitatively and blinded manner 0 = no staining; 1 = weak staining; 2 = moderate staining; and 3 = intense staining as described by us (87-89). E Paraffin-embedded lung tissue sections from paired WT and SPC-TAT mice were stained with H&E (image taken at 10x) for Lm analysis by MetaMorphTM. The mean linear intercept shows increased airspace enlargement. Data were shown as mean of n = 2–3/group  ± SEM; *significant (p < 0.05).

Full size image

Single-cell RNA sequencing of SPC-TAT mouse lungs

To investigate early events, pathways, and cell types involved in HIV TAT-associated lung inflammation as a function of transcriptomic changes, single-cell RNA sequencing (ScRNAseq) was performed on lungs from 4-month-old (juvenile) SP-C TAT mice and age-matched wild-type controls. Following quality control and filtering, we obtained high-quality single cells across all samples. Cell clustering and annotation revealed the major cell types: AT2 cells, multiciliated cells, adventitial fibroblasts, endothelial cells, macrophages, and T-cells. To account for batch effects, sample integration was performed using the Harmony algorithm, resulting in a unified UMAP representation. Figure 9A is a UMAP visualization identifying major immune, epithelial, endothelial, and stromal populations. Integration was performed on a merged Seurat object that had undergone normalization, identification of highly variable features, and principal component analysis. Key annotated clusters include AT2, AT1, multiciliated cells, alveolar macrophages, fibroblast subtypes, and diverse immune subsets. We observed SIRT1 suppression in AT2 cells in SP-C Tat mice (adjusted p-value of 0.012) (Fig. 9B). In line with our earlier data on suppression of CFTR by HIV TAT, we observed CFTR suppression in ciliated epithelial cells (adjusted p-value of 0.004) of SP-C TAT mice24,29 (Fig. 9C). Ciliated cells in the airway epithelium also demonstrated a non-significant reduction of core clock genes including REV-ERBα (NR1D1), BMAL1, PER1, CRY1, and RORA (Fig. 9C). Surprisingly, SIRT1 and PER2 levels were mildly elevated in ciliated cells (not significant), suggesting a possible compensatory mechanism in the early stages of these mice. Given that NR1D1 regulates metabolism and SIRT1 expression and activity are regulated by metabolism, decreased NR1D1 could lead to the small compensatory increase in SIRT1 levels.we did not observe secretion of proinflammatory cytokines in airway epithelial cells. Elevated expression of chemokines and proinflammatory cytokines was observed in adventitial fibroblasts (Fig. 9D), suggesting that lung fibroblasts may play an early role in inflammation in PLWH. Specifically, we found a considerable upregulation of CCL2 (MCP-1) in adventitial fibroblasts (adjusted p-values of 0.007). We also observed increased inflammation in CD4 T-cells, with a significant upregulation of IL5—in SP-C TAT mice (adjusted p-values of 0.0023) (Fig. 9E). The monocyte derived macrophages shows upregulation of IL13 (Fig. 9F). These data suggest that pulmonary fibroblasts, T-cells and macrophages may play an important initial role in lung inflammation in PLWH.

Fig. 9: ScRNA-Seq analysis of SPC TAT mice confirmed suppressed clock gene expression and increased proinflammatory gene expression.

A The UMAP was generated after sample integration using Harmony81. Harmony utilized a merged Seurat object as input, which encompassed data that had undergone normalization and included pre-defined highly variable features and principal components. The process of preparing this merged object entailed several steps: first, sctransform v2 normalization was performed independently on each sample, and integration features were identified across samples using Seurat; these individual normalized Seurat objects, designating variable features as integration features, were then merged and ultimately, 20 principal components on this merged object were calculated. Following the execution of the Harmony algorithm, we proceeded to perform RunUMAP with the parameter settings reduction = “harmony”, dims =1:20 to create a UMAP representation based on the Harmony embeddings. Subsequently, we conducted Louvain clustering utilizing the FindNeighbors function reduction = “harmony”, dims =1:20, and the FindClusters function with a resolution of 0.5, using the Seurat R program82. B Volcano plot for differential expression of genes in AT2 cells under TAT and control conditions, represented as log2 fold change. C Volcano plot for differential expression of genes in multiciliated (non-nasal) cells under TAT and control conditions, represented as log2 fold change. D Volcano plot for differential expression of genes in adventitial fibroblast cells under TAT and control conditions, represented as log2 fold change. E Volcano plot for differential expression of genes in CD4 T cells under TAT and control conditions, represented as log2 fold change. F Volcano plot for differential expression of genes in monocyte derived macrophages cells under TAT and control conditions, represented as log2 fold change. Horizontal line is draw at adjusted p-value < 0.05. Vertical line is draw at |log2FC| > 1.

Full size image

We and others have shown that circadian dysregulation in the lung can manifest as chronic lung inflammation18,54,55. Chromic lung inflammation is a hallmark of COPD56. COPD is one of the principal comorbidities in PLWH, with increased exacerbations and worse outcomes in HIV-positive smokers. Several studies have confirmed that the prevalence of COPD is higher in PLWH. PLWH often experience disturbed sleep and motor-cognitive issues due to circadian rhythm disorder57. HIV TAT disrupts the circadian system, affecting body temperature, hormone secretion, and immune cell circulation. Studies show that HIV TAT in the brain alters the sleep cycle. However, the molecular mechanisms underlying this effect, specifically the direct effects of HIV proteins on peripheral clocks in organs that have the potential to promote inflammation, have yet to be determined.

Our studies with NHBE cells, HIV-positive donor lungs, and SP-C TAT mice demonstrate for the first time that HIV infection dysregulates the molecular components of the lung circadian clock by an aberrant TAT-induced microRNAome. This microRNAome exhibits decreased SIRT1 levels, with consequent effects on the lung molecular clock genes. The circadian clock in the lungs, as well as the airway epithelium, plays an important role in maintaining inflammation homeostasis. We have demonstrated that disruption of this epithelial circadian clock is associated with lung inflammation and COPD18. To our knowledge, this is the first report that demonstrates the molecular mechanisms underlying HIV TAT-mediated lung inflammation, dysregulation of the lung circadian clock, and COPD development in PLWH.

Previous studies, including ours, have demonstrated that SIRT1 is the master circadian deacetylase and controls acetylation status of circadian genes, consequently affecting their activity as well as expression of the downstream molecular signaling cascade18,58,59. Given previous reports of HIV (and TAT) dysregulating the central clock and affecting sleep patterns, we first examined lungs from PLWH donors for changes in the expression of circadian genes and SIRT1 deacetylase. Surprisingly, we observed suppressed RNA levels of almost all core circadian genes as well as SIRT1. Next, we examined whether HIV infection dysregulates circadian oscillations of BMAL1 and PER2, two clock genes that are targets of SIRT1 deacetylase activity. Infection of NHBE ALI cells with HIV in culture demonstrated a decrease in levels of clock genes and the amplitude of circadian oscillations. Recombinant HIV TAT likewise demonstrated suppressed RNA levels of all circadian genes, and oscillation as well as protein levels of the SIRT1 targets BMAL1 and PER2. While previous reports have shown that HIV TAT can inhibit SIRT1 activity28,60, none of the reports showed that HIV TAT can also suppress SIRT1 biogenesis. Given that SIRT1 is a master deacetylase of the circadian genes and that SIRT1 suppression is associated with suppression of other clock genes leading to lung inflammation18, we examined the mechanism by which HIV TAT suppresses SIRT1. Our data demonstrated that HIV TAT-mediated SIRT1 suppression could be rescued by blocking microRNA processing. We previously reported that HIV TAT dysregulates the bronchial epithelial microRNAome29. We used miRNA-target algorithms to identify miR-126-3p as a potential regulator of SIRT1. In our earlier screen of the aberrant microRNAome by Dutta et al.29, this microRNA upregulation was borderline not significant (p = 0.0595). Using specific probes, we validated the statistically significant upregulation of this microRNA in NHBE ALI cultures treated with recombinant TAT as well as in lungs from HIV-positive donors. The miRNA-target relationship for miR-126-3p and SIRT1 was validated by demonstrating SIRT1 suppression in BEAS-2B airway epithelial cells transfected with miR-126-3p mimic. In our earlier studies, we also demonstrated that HIV upregulates miR-34a-5p as a microRNA that regulates SIRT161. In this study, we show that miR-126-3p, another microRNA upregulated by HIV TAT, also suppresses SIRT1. To the best of our knowledge, this is the first report that demonstrates miR-126-3p-mediated SIRT1 regulation. Our data showing that miR-126-3p alone can alter circadian gene expression and induce proinflammatory cytokines suggest that the actions of miR-34a-5p and miR-126-3p can regulate SIRT1 independently of each other. Our data also show that miR-126-3p mimics demonstrate a greater degree of SIRT1 suppression compared to miR-34a-5p mimic. The experiments required to disentangle the cooperative or independent effects of miR-34a-5p and miR-126a-3p are beyond the scope of this manuscript.

We used multiple experimental approaches to establish the relationship between TAT-mediated SIRT1 suppression and circadian dysregulation. siRNA-mediated SIRT1 suppression led to a corresponding increase in acetylated-BMAL1, and PER2-luciferase reporter assays demonstrated that TAT, as well as miR-126-3p, decreased the levels and amplitude of luciferase reporter oscillations. HIV TAT-mediated SIRT1 suppression and miR-126-3p mimic transfection led to increased secretion of proinflammatory cytokines. While we observed more cytokines upregulated by HIV TAT than miR-126-6p, this may be due to the model system used (NHBE for TAT versus BEAS-2B cells for miR-126-3p mimic). Alternatively, it is possible that TAT has dual effects on SIRT1 in that it suppresses SIRT1 biogenesis (mRNA levels) as well as activity. In contrast, miR-126-3p suppresses only SIRT1 biogenesis, leading to a more robust downregulation of SIRT1 effects with TAT than with miR-126-3p. Of note, MCP-1 (CCL2) and IFN-γ were induced by both TAT and miR-126-3p. We observed that treatment with SRT1720 completely reversed expression of the proinflammatory cytokines IL-6, GM-CSF, IP-10, MCP-1, IL-8, and G-CSF induced by HIV infection. We also observed a general decrease in baseline levels of several key proinflammatory cytokines.

Finally, we validated the effects of HIV TAT in 7-month-old SP-C TAT transgenic mice in which TAT expression is restricted to the airways. We observed an identical pattern of suppression in the circadian genes in lungs from SP-C TAT transgenic mice to that we observed with HIV-positive human donor lungs, with statistically significant suppression of BMAL1, CRY1, CRY2, PER1, PER2, REV-ERBα (NR1D1), CLOCK, RORA and SIRT1. Histology experiments with paired lungs showed that SP-C TAT mice demonstrated decreased SIRT1, PER2, and REV-ERB α expression, and increased oxidative burden (4HNE staining; Fig. 8D) and airspace enlargement characteristic of emphysema. Our data demonstrate that HIV TAT, and by extension, HIV infection, induces miR-126-3p to suppress SIRT1. SIRT1 suppression then dysregulates the core lung circadian clock which then manifests as suppression of other clock genes, including REV-ERBα, leading to inflammation and features of emphysema. While the exact mechanism by which clock dysregulation leads to inflammation remains to be determined, studies have shown that REV-ERBα suppression leads to inflammation in multiple tissues. Another mechanism may be a direct effect of BMAL1 suppression. The CLOCK gene is a positive regulator of, and associates with NFκβ62. SIRT1-mediated BMAL1 deacetylation counteracts CLOCK-dependent NF-κβ activation62. Alternatively, SIRT1 can directly deacetylate p65 to inhibit NF-κβ63. Hence, a combination of SIRT1 suppression and circadian dysregulation can lead to increased NF-κβ activity and decreased REV-ERBα levels, resulting in lung inflammation.

COPD is a multifactorial disease, with multiple interconnected pathways involved in lung inflammation and COPD development. Our data demonstrate that SIRT1 suppression and circadian dysregulation in the lung play an important role in the inflammatory process. PLWH present with emphysema as early as 38–47 years51,52. Understanding early events and cell types involved in lung inflammation can help arrest the development of emphysema. We focused on early transcriptomic changes that may point to initial pathways and cell types involved in lung inflammation, carrying out scRNAseq experiments on ~4-month-old lungs (human age equivalent ~20–25-year-old53). The scRNAseq experiment confirmed the suppression of SIRT1 in AT2 cells in 4-month old mice. ScRNAseq data also validated the HIV TAT-mediated CFTR suppression in epithelial cells previously reported by us29, and identified that this suppression occurs in ciliated cells of the airway epithelium (Fig. 9C). We also observed suppression of clock genes, including REV-ERBα (NR1D1), in ciliated cells. However, at this early age, the suppression had not reached statistical significance yet (not shown). We saw several other cell types driving inflammation in SPC-Tat mice. We observed elevated expression of chemokines and proinflammatory cytokines in adventitial fibroblasts, suggesting that ciliated cells and lung fibroblasts may be early drivers of lung inflammation in PLWH. Of note, we found elevated CCL2 (MCP-1) expression in adventitial fibroblasts in TAT mice. CCL2 is increased in emphysema and is produced in response to inflammatory stimuli by a variety of cells64. Higher CCL2 induces an inflammatory cytokine response and pulmonary leukocyte accumulation. CCL2 increase is also associated with other lung diseases such as pulmonary hypertension, acute lung injury, and pulmonary fibrosis65,66. CCL2 is known to activate CCR2 on monocytes and T cells67. T-lymphocytes are the predominant inflammatory cells present in the central airways and lung parenchyma of COPD patients, and their presence correlates with the degree of alveolar damage and the severity of airway obstruction in these individuals68. CD4+ T cells show significant upregulation of IL5. IL5 plays an important role in driving airway eosinophilic inflammation in both asthma and COPD69. Monocyte-derived macrophages demonstrate upregulation of IL13 which promotes mucus cell hyperplasia70, and airway remodeling by driving fibroblast proliferation71. Identifying these multiple pathways leading to COPD pathology and their links to lung circadian dysregulation is beyond the scope of this manuscript.

A recent concise review by Shafaati et al.72 examined the interplay between HIV infection or HIV-encoded proteins and host circadian timing systems, including molecular clock regulators such as CLOCK. BMAL1, PER/CRY, and REV-ERBα, and discussed how HIV may perturb sleep-wake cycles and antioxidant defenses72. However, our data identifies a mechanism by which HIV dysregulate peripheral clocks (in our case, lungs) leading to inflammation. We further believe that a similar mechanism may underlie dysregulation of the central SCN clock, thereby linking our findings to the broader systemic effects proposed in that work. In conclusion, our data demonstrate for the first time that HIV TAT dysregulates the lung molecular clock to promote inflammation. Further understanding of the downstream pathways can help identify molecular targets to mitigate lung inflammation. Resetting the lung molecular clock using REV-ERBα and SIRT1 activators can serve as therapeutic leads to mitigate lung inflammation, thereby arresting the development and progression of emphysema.

Methods

Data availability