HIV-1 and SIV Infection Are Associated with Early Loss of Lung Interstitial CD4+ T Cells and Dissemination of Pulmonary Tuberculosis

Abstract

Lung interstitial CD4+ T cells are critical for protection against pulmonary infections, but the fate of this population during HIV-1 infection is not well described. We studied CD4+ T cells in the setting of HIV-1 infection in human lung tissue, humanized mice, and a Mycobacterium tuberculosis (Mtb)/simian immunodeficiency virus (SIV) nonhuman primate co-infection model. Infection with a CCR5-tropic strain of HIV-1 or SIV results in severe and rapid loss of lung interstitial CD4+ T cells but not blood or lung alveolar CD4+ T cells. This is accompanied by high HIV-1 production in these cells in vitro and in vivo. Importantly, during early SIV infection, loss of lung interstitial CD4+ T cells is associated with increased dissemination of pulmonary Mtb infection. We show that lung interstitial CD4+ T cells serve as an efficient target for HIV-1 and SIV infection that leads to their early depletion and an increased risk of disseminated tuberculosis.

Keywords: BAL; CD4+ T cells; HIV-1; HIV/TB co-infection; cell death; lung; tuberculosis.

Figure 1.. CCR5-Tropic HIV-1 Infection Induced Severe Depletion of Human Lung CD4+ T Cells

Single-cell suspensions were obtained from human lung, tonsil, and blood samples. (A and B) The frequency of (A) T_RM-like CD4+ cells (TCRα/β+CD45RO+CD62L−CD25−CD69+) and (B) HIV-1 co-receptor CCR5+ memory CD4+ T cells was determined by flow cytometry. (C–F) 0.5 × 10⁶ cells were cultured in a V-bottom 96-well plate, and, where indicated, cells were incubated with antiretroviral (ARV) drugs (DRV, darunavir [a protease inhibitor]; RAL, raltegravir [an integrase inhibitor]; AZT, zidovudine [a reverse transcriptase (RT) inhibitor]; EFV, efavirenz [an RT inhibitor]; MVC, maraviroc [a CCR5 antagonist]) before infection with CCR5-tropic NL4-3 GFP HIV-1 and analyzed by flow cytometry. (C) The percentage of viable CD4+ T cells was determined relative to untreated cells. (D) The percentage of productively infected CD4+ T cells was determined by analyzing HIV-1 GFP+ CD4+ T cells. (E) Correlation between percentage of viable CD4+ T cell and productively infected cells. (F) Percentage of viable CD4+ T cells in mock-infected, HIV-1-infected, and HIV-1-infected samples pre-incubated with different ARVs. The p values were measured by (A–D and F) Kruskal-Wallis and Dunn’s multiple comparisons tests or (E) by Spearman r test. Scatter plots are labeled with median and interquartile range. Each data point represents the average of duplicates from one subject. See also Figure S1.

Figure 2.. HIV-1 Infection Results in Severe Depletion of Lung Interstitial CD4+ T Cells In Vivo

Humanized mice from 2 different batches were intravaginally infected with 50,000 infectious particles of HIV-1 (JR-CSF) (n = 14) or left uninfected as controls (n = 13). (A–D) 4 and 7 weeks after infection, cells from (A) the lung interstitium, (B) BAL, (C) the spleen, and (D) the blood were analyzed for CD4+ T cell loss by comparing CD4+ T cell numbers with uninfected control animals using flow cytometry. (E and F) CD4+ T cell loss in paired BAL and lung samples was analyzed as fold change compared with the median CD4+ T cell count of uninfected animals (E) 4 weeks or (F) 7 weeks after infection. (G) Depletion of CD4+ T cells 7 weeks after infection in lungs compared with the spleen was confirmed by immunohistochemistry (IHC) staining for CD4. (H and I) CD4+ cells in (H) the lungs and (I) the spleen were quantified using Histoquest software. The p values were measured by Kruskal-Wallis and Dunn’s multiple comparisonstests and (H and I) Mann-Whitney U test. Scatterplots are labeled with median and interquartile range. Each data point represents one humanized mouse sample. See also Figure S2.

Figure 3.. CD4+ T Cells in the Lungs Are Highly Susceptible to Productive HIV-1 Infection In Vivo

(A and B) 7 weeks after infection, RNA was extracted from lung and spleen (A) total tissue or (B) from sorted CD4+ T cells from the lungs and spleen. HIV-1 RNA was detected by HIV-1 gag qPCR and quantified by using a linear HIV-1 standard and normalized to CD4+ T cell counts. (C and D) Production of virus in lung CD4+ cells was confirmed by protein staining of HIV-1 p24 using IHC (C). The ratio of p24+ cells:CD4+ cells in the lungs and spleen was quantified using Histoquest software (D). The p values were measured by (A and D) Mann-Whitney U test and (B) Kruskal-Wallis and Dunn’s multiple comparisons tests. Scatterplots are labeled with median and interquartile range. Each data point represents one humanized mouse sample.

Figure 4.. CD69+ Memory CD45iv− Lung Interstitial CD4+ T Cells Are Most Significantly Affected by HIV-1-Induced CD4+ T Cell Loss

HIV-1-infected humanized mice (n = 6) or uninfected control mice (n = 12) were labeled by injection of an anti-human CD45 antibody for 3 min prior to sacrifice. CD45iv− and positive CD4+ T cells were analyzed in the lungs, BAL, spleen, and blood by flow cytometry. (A) Representative plots of CD45iv+ and CD45iv− CD4+ T cells in the lungs, BAL, spleen, and blood. (B–D) The T_RM-like phenotype of CD45iv+ or CD45iv− CD4+ T cells was further characterized by flow cytometry. (B) Fold lung CD4+ T cell loss was analyzed by comparing CD4+ T cell numbers from infected with median uninfected control animals. (C) The percentage of all CD4+ T cells with surface expression of CD69+ was analyzed by gating on viable CD4+ TCRα/β+CD45RO+CD62L−CD25− T cells. (D) The percentage of all CD4+ T cells with surface expression of CCR5+ was analyzed by gating on viable CD4+ T cells. (E and F) Correlation between intracellular HIV-1 p24+ CD4+ T cells 4 weeks after infection and number of (E) lung CD45iv− CD4+ T cell and (F) blood CD45iv+ CD4+ T cells. The p values were measured by (A−D) Kruskal-Wallis and Dunn’s multiple comparisons tests and (E and F) Spearman test. Scatterplots are labeled with median and interquartile range. Each data point represents one humanized mouse sample. See also Figures S3 and S4.

Figure 5.. Mtb/SIV Co-infection in NHPs Leads to Severe CD4+ T Cell Depletion in the Lung Interstitium, which Is Associated with Increased Disseminated TB

Rhesus macaques were infected with Mtb CDC1551 via low-dose aerosol challenge to establish LTBI (n = 26). 9 weeks after Mtb infection, a subset ofthe NHPs (n = 15) was co-infected with SIVmac239. Necropsy was performed 20–22 weeks after Mtb infection or after TB reactivation with collection of the lungs, spleen, kidneys, and liver. BAL was performed prior to necropsy. (A–C) None of the SIV-uninfected animals progressed from LTBI (circles) to active tuberculosis (ATB; triangles), whereas 8 animals in the SIV-infected group developed ATB. CD4+ T cell numbers in lung tissue (A), BAL (B), and blood (C) were assessed by flow cytometry. (D) CD4+ T cell loss relative to the median in the CD4+ T cell count in SIV-uninfected NHPs in paired BAL and lung samples was determined. (E) Mtb burden in the liver was measured in SIV-infected and uninfected NHPs, and lung interstitial CD4+ T cell numbers were correlated with liver CFUs/g tissue. The p values were measured by (A–C and E) Mann-Whitney U test, (D) Wilcoxon test, or (F) Spearman r test. Scatterplots are labeled with median and interquartile range. Each data point represents one NHP. See also Figure S5.