Inhibition of telomerase activity alters tight junction protein expression and induces transendothelial migration of HIV-1-infected cells

Abstract

Telomerase, via its catalytic component telomerase reverse transcriptase (TERT), extends telomeres of eukaryotic chromosomes. The importance of this reaction is related to the fact that telomere shortening is a rate-limiting mechanism for human life span that induces cell senescence and contributes to the development of age-related pathologies. The aim of the present study was to evaluate whether the modulation of telomerase activity can influence human immunodeficiency virus type 1 (HIV-1)-mediated dysfunction of human brain endothelial cells (hCMEC/D3 cells) and transendothelial migration of HIV-1-infected cells. Telomerase activity was modulated in hCMEC/D3 cells via small interfering RNA-targeting human TERT (hTERT) or by using a specific pharmacological inhibitor of telomerase, TAG-6. The inhibition of hTERT resulted in the upregulation of HIV-1-induced overexpression of intercellular adhesion molecule-1 via the nuclear factor-κB-regulated mechanism and induced the transendothelial migration of HIV-1-infected monocytic U937 cells. In addition, the blocking of hTERT activity potentiated a HIV-induced downregulation of the expression of tight junction proteins. These results were confirmed in TERT-deficient mice injected with HIV-1-specific protein Tat into the cerebral vasculature. Further studies revealed that the upregulation of matrix metalloproteinase-9 is the underlying mechanisms of disruption of tight junction proteins in hCMEC/D3 cells with inhibited TERT and exposed to HIV-1. These results indicate that the senescence of brain endothelial cells may predispose to the HIV-induced upregulation of inflammatory mediators and the disruption of the barrier function at the level of the brain endothelium.

telomeres are nucleoprotein structures that protect the end of linear chromosomes from DNA degradation (6). The length of the telomeres progressively shortens with each cell division in somatic cells because of incomplete DNA replication (30, 41). Functional telomeres are essential for cell proliferation; therefore, cells undergo replicative senescence and growth arrest when telomeres become short (40, 41). It has been proposed that telomere shortening is a rate-limiting mechanism for human life span that also contributes to the development of age-related pathologies (41). Indeed, individuals with shorter telomeres are prone to develop degenerative diseases that occur during human aging, including heart failure and atherosclerosis (7, 20, 22). In the context of human immunodeficiency virus type 1 (HIV-1) infection, it was reported that HIV-1 or HIV-1 protein Tat can downregulate telomerase activity and human telomerase reverse transcriptase (hTERT) expression in peripheral blood mononuclear cells and the nucleus of CD4⁺ T lymphocytes (19).

The functional telomerase complex contains a catalytic rate-limiting component, TERT, and a telomerase template RNA. Telomerase expression is suppressed in the majority of somatic cells, including endothelial cells (52). However, TERT mRNA can be detected in specific endothelial cell populations, such as endothelial cells from astrocytic tumors (34). Progressive telomere shortening has been directly implicated in endothelial senescence and endothelial cell biology. For example, there was a significant loss of telomere length in endothelial cells from iliac arteries compared with iliac veins, an observation that is consistent with the increased cell turnover in arteries (12). In the present study, we hypothesize that the modulation of telomerase length in endothelial cells may influence vascular responses induced by HIV-1. This hypothesis will be studied in the context of the blood-brain barrier (BBB) properties, because HIV-1 affects the integrity of the BBB early in the infection process.

Antiretroviral therapy increased survival rates of HIV-1-infected individuals. As a result, a significant percentage of patients are now over 50 yr old. In addition, there is a rapid growth of new infections in older people (27). Older people with HIV-1 infection have a significantly higher prevalence of neurodegenerative diseases, including dementia, compared with younger patients (48). Several factors may contribute to this phenomenon (8), including vascular pathology, age-related immunological changes, and limited compensatory brain capacity. However, there are no reports linking HIV-1-mediated alterations in the brain vasculature to telomere shortening.

With the knowledge that the loss of telomere length results in cellular senescence and that individuals with short telomeres are prone to the development of neurodegenerative diseases, the aim of the present study was to evaluate whether the modulation of telomerase activity can influence HIV-1-mediated dysfunction of brain endothelial cells. Our data indicate that the inhibition of hTERT can stimulate the senescence of cultured brain endothelial cells, induce inflammatory responses, and lead to the disruption of tight junction protein expression via increased matrix metalloproteinase-9 (MMP-9) activity.

MATERIALS AND METHODS

Cell cultures, generation of HIV-1 stock, and HIV-1 treatment.

The hCMEC/D3 cell line was recently developed based on the overexpression of hTERT in human brain endothelial cells (51). hCMEC/D3 cells were cultured in endothelial cell basal medium-2 (EBM-2; Cambrex BioScience, Walkersville, MD) as described previously (3, 26). Human monocytic U937 cells were cultured in RPMI-1640 medium (Invitrogen, Carlsbad, CA), supplemented with 10% FBS and antibiotics (penicillin, 100 U/ml; and streptomycin, 100 μg/ml, Invitrogen).

HIV-1 stock was generated by the transfecting of human embryonic kidney 293T cells (American Type Culture Collection, Manassas, VA) with the HIV-1 PYK-JRCSF plasmid carrying 0.5 kb of 3′-flanking sequences and 2.2 kb of 5′-flanking DNA (10). p24 antigen was determined by ELISA (ZeptoMetrix, Buffalo, NY) as described earlier (26) as a marker of HIV-1 infection. HIV-1 stock was then used to infect U937 cells. Briefly, 5 × 10⁶ U937 cells in T-25 flasks (Corning, NY) were infected with viral isolate containing 120 ng of p24 in a final volume of 5 ml. After an overnight incubation, the cells were resuspended in fresh medium and maintained for an additional 3–5 days for viral replication. For the majority of the experiments, confluent hCMEC/D3 cells cultured on six-well Transwell plates (pore size, 0.4 μm, Corning Costar, Corning, NY) were exposed to 1.5 × 10⁶ HIV-infected or uninfected monocytes per well added to the top chamber.

TERT-deficient mice and isolation of brain microvessels.

TERT^−/− mice were generated on the C57BL/6J genetic background and bred through heterozygous mating (14). All experimental procedures and protocols were reviewed and approved by the Animal Care and Use Committee of the University of Kentucky. Studies were conducted on 6-mo-old female mice. Vehicle or HIV-1-specific Tat protein (50 μg/mouse) was administered twice in 12-h intervals into the internal carotid artery (ICA) using our previously described technique (13). Brain microvessels were isolated 24 h after the first injection. Briefly, mice were anesthetized and perfused as described earlier (37). Following decapitation, the brains were removed and immediately immersed in ice-cold isolation buffer with Complete Protease Inhibitor (Roche; Indianapolis, IN). The choroid plexus, meninges, cerebellum, and brain stem were removed; the brains were homogenized in isolation buffer with Complete Protease Inhibitor, followed by dextran gradient centrifugation. The obtained pellets containing microvessels were either smeared on slides for confocal analysis or resuspended in 0.5 ml of 6 M urea lysis buffer containing 6 M urea, 0.1% Triton X-100, 10 mM Tris (pH 8.0), 5 mM MgCl₂, 5 mM EDTA, and 150 mM NaCl with Complete Protease Inhibitor for Western blot analyses. Protein samples were either immediately used or stored at −80°C (39).

hTERT silencing and senescence-associated β-galactosidase activity assay.

Small interfering RNA (siRNA) oligomers specifically targeting hTERT were obtained from Dharmacon (Chicago, IL). Transfections were performed with 40 nM anti-hTERT or control siRNA for 5 h using the GeneSilencer (Genlantis, San Diego, CA) (25, 56). Following transfections, the cells were allowed to recover in complete medium for 24 h, followed by a treatment with HIV-infected or control monocytes on the next day. In selected experiments, TERT activity was inhibited by a 2-h pretreatment with telomerase inhibitor TAG-6 (2 μM; Calbiochem, San Diego, CA).

Staining for β-galactosidase activity was used as a marker of cellular senescence (17). The assay was performed using the Senescence β-Galactosidase Staining Kit (Cell Signaling Technology, Danvers, MA) according to the protocol provided by the manufacturer.

Telomeric repeat amplification protocol assay.

Telomeric repeat amplification protocol (TRAP) assay was performed using the TRAPeze XL telomerase detection kit (Chemicon International, Temecula, CA) according to the manufacturer's instructions. PCR products were measured using a fluorescence plate reader (SPECTRAmax GEMINI XS, Molecular Devices, Sunnyvale, CA). Heat-inactivated protein extracts were used as negative controls, and protein extracts from telomerase-positive cells provided in the kit were used as positive controls.

Western blot analysis and immunofluorescence microscopy.

Protein was extracted using the radioimmunoprecipitation assay lysis buffer (Santa Cruz Biotechnology, Santa Cruz, CA) and centrifuged at 15,000 g for 15 min. The supernatants were collected and protein concentrations were determined using the BCA Protein Assay Kit (Pierce, Rockford, IL). Samples were separated on 4–15% Tris·HCl Ready SDS-polyacrylamide gel (Bio-Rad, Hercules, CA). Anti-hTERT antibody was obtained from Calbiochem (Gibbstown, NJ), and anti-nuclear factor-κB (NF-κB) p65 antibody was from Santa Cruz Biotechnology. Anti-zonula occludens-1 (ZO-1) and claudin-5 antibodies were purchased from Zymed (San Francisco, CA), and anti-intercellular adhesion molecule (ICAM-1) antibody was from BD Transduction (Franklin Lakes, NJ). Anti-actin antibody was purchased from Sigma (St. Louis, MO), and all secondary antibodies were from Santa Cruz Biotechnology.

Immunofluorescence microscopy.

For immunofluorescence microscopy, brain microvessels smeared on slides were fixed for 10 min at 95°C, followed by an incubation with 3% formaldehyde in phosphate-buffered saline (PBS) for 10 min at 25°C. The slides were washed five times with PBS, permeabilized with 0.1% Triton X-100 for 30 min, rewashed five times in PBS, and blocked in 1% bovine serum albumin (BSA) in PBS for 30 min at 25°C. Samples were then incubated with an individual primary antibody (1:500 dilutions in 1% BSA in PBS) overnight at 4°C. The slides were washed with PBS and incubated with Alexafluor 568-conjugated anti-mouse IgG or Alexafluor 488-conjugated anti-rabbit IgG (Invitrogen; 1:1,000 dilution in 1% BSA in PBS) for 1 h at 37°C. After a final washing with PBS, the slides were mounted with ProLong Gold Antifade reagent containing 4′,6-diamidino-2-phenylindole (Invitrogen) to visualize the nuclei. The immunofluorescent images were evaluated and captured using confocal microscopy.

Transendothelial migration assay.

Transendothelial migration was measured as described earlier (25). Cocultures of hCMEC/D3 were generated on the opposite sides of inserts of the 12-well Transwell system (pore size, 3 μm; Corning Costar, Corning, NY). Astrocytes (1 × 10⁵ cells) were plated on the lower surface of the membrane and allowed to attach for 5 h. The inserts were then placed upright, and 1 × 10⁵ hCMEC/D3 cells were plated onto the upper site of the membrane. The upper chamber contained complete EBM-2 medium suitable for hCMEC/D3, and the lower chamber contained DMEM containing 10% FBS and antibiotics (100 U/ml penicillin and 100 μg/ml streptomycin) to support astrocyte growth.

HIV-infected or uninfected U937 cells were labeled with 5 μM calcein-AM (Calbiochem, San Diego, CA). Then, 1 × 10⁶ labeled cells were added to the top chamber of the Transwell system, and transmigration was allowed for 24 h in a cell culture incubator. Fluorescence was measured in aliquots of 100 μl collected from the lower chamber at 480 nm for excitation and 530 nm for emission.

NF-κB studies.

NF-κB transactivation assay was performed as described previously (26). Briefly, hCMEC/D3 cells cultured on 12-well plates were transfected with 0.5 μg of NF-κB-responsive reporter construct (pNF-κB-Luc; Stratagene, La Jolla, CA) and cotransfected with 0.05 μg of the control pRL-TK construct (Promega, Madison, WI). The transfection procedures were performed for 5 h using lipofectin (Invitrogen). Following transfections, the cells were washed and allowed to recover for 12 h in normal growth medium, pretreated with TAG-6, and then directly exposed to HIV-1 particles (200 pg/ml of p24). Cocultures with HIV-infected cells were not used in these experiments because of potential anti-HIV effects of telomerase inhibitors. Firefly and renilla luciferase activities were determined using the Dual-Luciferase Reporter Assay System (Promega) (2, 26).

The effects of telomerase inhibition on the NF-κB translocation were confirmed by assessing the levels of NF-κB p65 in nuclear extracts of the treated cells by Western blot analysis. Nuclear extracts were prepared using NucBuster Protein Extraction Kit (Novagen, San Diego, CA). In selected experiments, NF-κB activity was inhibited by a specific inhibitor, 4-methyl-N1-(3-phenylpropyl)benzene-1,2-diamine (JSH-23), purchased from Santa Cruz Biotechnology.

MMP-9 promoter activity assay and MMP-9 activity assay.

To generate firefly luciferase reporter constructs of MMP-9 (pGL3 MMP-9), the 5′ -flanking region of human MMP-9 (−1,729 to +3) was amplified by PCR from human genomic DNA. The fragment was then cloned to the pGL3-Basic vector (Promega) by inserting between MluI and NcoI sites. In addition, NF-κB binding site 1 (−630 to −610) 5′-GGAATTCCCC-3′ was mutated into 5′-GCAATTCCCC-3′, and NF-κB binding site 2 (−350 to −340) 5′-GGGGATCCC-3′ was mutated into 5′-GGGCGATCCC-3′ (boldface letters indicate the mutated or inserted bases). The mutations were verified by DNA sequencing after subcloning the mutated MMP-9 promoters into firefly luciferase basic vector pGL3. hCMEC/D3 cells were transfected with 0.5 μg wild-type (pGL3 MMP-9) or mutated (pGL3 mt-MMP-9) MMP-9 promoter constructs using lipofectin (Invitrogen) as a transfection reagent. To normalize for transfection efficiency, the cells were cotransfected with 0.05 μg pRL-TK construct (Promega) encoding for renilla luciferase. Firefly and renilla luciferase activities were determined using the Dual-Luciferase Reporter Assay System (Promega) (2, 26).

MMP-9 activity was detected by gelatin zymography (25, 35) on premade 10% polyacrylamide gels containing 0.1% gelatin (Invitrogen) using 10-μl aliquots of serum-free media from treated cultures.

Statistical analysis.

Routine statistical analysis was completed using SigmaStat 2.03 (SPSS, Chicago, IL). One- or two-way ANOVA was used to compare mean responses among the treatments. Statistical probability of P < 0.05 was considered significant.

RESULTS

hTERT silencing stimulates senescence of hCMEC/D3 cells.

A gene silencing technology was employed to decrease the activity of TERT in hCMEC/D3 cells. As illustrated in Fig. 1A, the transfection of hCMEC/D3 cells with siRNA targeted against hTERT resulted in ∼40% decrease in hTERT protein expression. The exposure to HIV-1-infected monocytic U937 cells also significantly diminished hTERT expression. However, the most significant decrease in hTERT levels was observed in cells with silenced hTERT and exposed to HIV-1-infected cells. A decrease in hTERT protein expression was confirmed by the TRAP assay. Quantification of the TRAP products by fluorescence measurements fully confirmed the lowest hTERT activity in hCMEC/D3 cells with silenced hTERT and exposed to HIV-1 infected U937 cells (Fig. 1B).

Fig. 1.

Human telomerase reverse transcriptase (hTERT) silencing results in senescence of hCMEC/D3 cells. Confluent hCMEC/D3 cells cultured on 6-well plates were exposed to 1.5 × 10⁶ HIV-1-infected monocytic U937 cells for 24 h. Controls were exposed to the same amount of uninfected U937 cells. Silencing was performed by transfection of hCMEC/D3 cells with hTERT-specific or control siRNA for 5 h. A: the effects of HIV-1 exposure and/or hTERT silencing on hTERT protein expression was analyzed by Western blot analysis. B: telomerase activity was measured by telomeric repeat amplification protocol assay, and the amplification products were detected by a fluorescence plate reader. TPG, total product generated. C: cellular senescence was analyzed by the assessment of β-galactosidase activity by histochemistry. Cells positive for β-galactosidase activity are stained blue (arrows). Results are means ± SE of 4 separate experiments. *P < 0.05 or **P < 0.01, data in cultures exposed to HIV-1 are significantly different from the corresponding controls without HIV treatment. †P < 0.05 or ††P < 0.01, data in control [nonexposed to human immunodeficiency virus type 1 (HIV-1)] cultures transfected with hTERT small interfering RNA (siRNA) are significantly different from the corresponding control cultures transfected with control siRNA. #P < 0.05, data in cultures transfected with hTERT siRNA and treated with HIV-1 are significantly different from the corresponding cultures transfected with control siRNA and exposed to HIV-1.

Positive staining for β-galactosidase at pH 6.0 is a widely used marker of cellular senescence. As indicated in Fig. 1C, β-galactosidase was histochemically detectable in cells with silenced TERT (arrows), indicating that a partial loss of TERT expression is connected with accelerated hCMEC/D3 senescence.

HIV-induced overexpression of ICAM-1 and transendothelial migration of monocytic cells is increased by inhibition of hTERT.

The stimulation of proinflammatory responses is the prominent feature of HIV-1-induced dysfunction of brain endothelial cells (9, 45). Therefore, we evaluated the effects of hTERT inhibition on the expression of ICAM-1, which is the primary mediator of leukocyte adhesion to the surface of the brain endothelium. As indicated in Fig. 2A, the exposure of hCMEC/D3 cells to HIV-1-infected U937 cells significantly increased mRNA levels of ICAM-1. Interestingly, this effect was significantly potentiated by the silencing of hTERT. Changes in mRNA levels were accompanied by alterations of ICAM-1 protein expression. Indeed, the HIV-1-mediated increase in ICAM-1 protein was potentiated by hTERT silencing compared with cells transfected with control siRNA (Fig. 2B). Similar effects were achieved by blocking TERT activity by TAG-6 (Fig. 2C).

Fig. 2.

Inhibition of hTERT potentiates HIV-induced overexpression of ICAM-1 in hCMEC/D3 cells. A and B: hCMEC/D3 cultures were transfected with hTERT or control siRNA and exposed to HIV-1-infected or uninfected monocytes as in Fig. 1. In selected experiments (C), cultures were pretreated with telomerase inhibitor, TAG-6 (2 μM), for 2 h instead of transfection with hTERT siRNA. ICAM-1 mRNA (A) and protein (B and C) were determined by real-time RT-PCR and Western blot analysis, respectively. The blots in B and C are representative images from 3 independent experiments, and the quantified results are depicted in the bar graphs. D: inhibition of hTERT potentiates HIV-induced monocyte migration across the in vitro model of the blood-brain barrier. The blood-brain barrier model consisted of cocultures of hCMEC/D3 cells with astrocytes plated in upper and lower chambers, respectively, of the Transwell system. hCMEC/D3 cells were transfected with hTERT or control siRNA as in Fig. 1. HIV-1-infected or uninfected U937 cells were labeled with calcein-AM, added to the upper chamber of the Transwell system in the amount of 1 × 10⁶, and allowed to migrate across hCMEC/D3 monolayers for 24 h. Results are means ± SE of 4 separate experiments. *P < 0.05 or **P < 0.01, data in cultures exposed to HIV-1 are significantly different from the corresponding controls without HIV treatment. †P < 0.05, data in cultures with inhibited TERT activity by hTERT siRNA or TAG-6 and not exposed to HIV are significantly different from the corresponding control cultures without TERT inhibition. #P < 0.05 or ##P < 0.01, data in cultures with inhibited TERT by hTERT siRNA or TAG-6 and treated with HIV-1 are significantly different from the corresponding cultures without TERT inhibition and exposed to HIV-1.

Exposure to HIV-1 can disrupt the barrier function of the brain endothelium and stimulate a migration of inflammatory cells into the brain (36, 45). Therefore, we evaluated how the downregulation of TERT activity in hCMEC/D3 cells can affect the transendothelial migration of HIV-1-infected monocytic U937 cells using a functional assay of transendothelial migration and the in vitro model of the BBB. As shown in Fig. 2D, the migration of HIV-1-infected U937 cells was significantly higher than that of uninfected cells. In addition, the silencing of hTERT significantly elevated the passage of HIV-infected U937 cells across the endothelial monolayers compared with the controls in which hCMEC/D3 cells were transfected with nonspecific siRNA and cocultured with noninfected U937 cells.

HIV-1 Tat-induced activation of ICAM-1 in brain capillaries is potentiated in TERT-deficient mice.

To confirm the involvement of TERT in HIV-1-mediated inflammatory responses, we next performed a series of animal studies on wild-type and TERT-deficient mice. HIV-1 is not infectious to mice; therefore, the recombinant HIV-1 protein Tat was used in all these experiments. The results from our (3, 26, 56) and other (50, 54) laboratories indicated that several vascular effects of HIV-1 can be reproduced by treatment with Tat. Tat was injected into the ICA, which allowed the administration of Tat directly into the brain vasculature, mimicking the pathology of HIV-1 in which brain capillaries are exposed to HIV-1 and HIV-1 proteins from their luminal site. The injection of Tat (50 μg) resulted in the upregulation of ICAM-1 expression as demonstrated by immunostaining in freshly isolated brain microvessels and quantified by densitometry analysis of Western blots (Fig. 3, A and B). Importantly, Tat-mediated upregulation of ICAM-1 was more pronounced in TERT-deficient mice compared with wild-type controls.

Fig. 3.

HIV Tat-induced overexpression of ICAM-1 is potentiated in brain microvessels of TERT-deficient (TERT^−/−) mice. Tat protein (50 μg/mouse) was administered into the internal carotid artery of TERT^−/− or wild-type (WT) mice. Brain microvessels were isolated 24 h post-Tat injection and analyzed for ICAM-1 using immunofluorescence (A) and Western blot analysis (B). Control mice were injected with vehicle. The immunostaining data are merged immunofluorescence and phase-contrast micrographs acquired using a ×60 oil-immersion lens under identical instrument settings. Western blot analyses were preformed on isolated microvessels. The blots are representative images from 3 mice per group, and the quantified results are depicted in the bar graphs. Results are means ± SE. *P < 0.05 or **P < 0.01, data in WT or TERT^−/− mice injected with Tat are significantly different from the corresponding controls injected with vehicle. ##P < 0.01, data in TERT^−/− mice treated with Tat are significantly different from the corresponding WT mice injected with Tat.

NF-κB regulates overexpression of ICAM-1 induced by HIV-1 and inhibition of hTERT.

NF-κB is the main transcription factor that regulates the expression of inflammatory genes, including adhesion molecules. Therefore, we evaluated the hypothesis that NF-κB is involved in alterations of ICAM-1 expression observed in hCMEC/D3 cells with decreased TERT activity and exposed to HIV-1. To determine the NF-κB transactivation, hCMEC/D3 cells were transfected with the pNF-κB-Luc construct containing five repeats of NF-κB enhancer element. The exposure of transfected cells to HIV-1 or telomerase inhibitor TAG-6 significantly increased NF-κB transactivation as determined by luciferase activity. However, the most significant increase was observed in cells that were exposed to both these factors simultaneously (Fig. 4A).

Fig. 4.

NF-κB regulates ICAM-1 overexpression induced by telomerase inhibition and HIV-1- exposure. A: NF-κB transactivation was determined by luciferase activity in hCMEC/D3 cells upon transfection with the pNF-κB-Luc construct and treatment with HIV-1 particles (200 pg p24/ml) for 24 h. Cotransfection with the control pRL-TK construct allowed normalization of the obtained results to renilla luciferase. Cells were pretreated with TAG-6 (2 μM) 2 h before HIV-1 exposure. B: NF-κB p65 levels were determined in nuclear fraction of hCMEC/D3 cells pretreated with TAG-6 (2 μM; 2 h before HIV-1 exposure) and exposed to 1.5 × 10⁶ HIV-1-infected or uninfected U937 cells for 24 h. C: ICAM-1 protein expression was determined by Western blot analysis in cells transfected with hTERT and exposed to HIV-1-infected U937 cells as in Fig. 1. Selected cultures were pretreated with 4-methyl-N1-(3-phenylpropyl)benzene-1,2-diamine (JSH-23) (30 μM) 2 h before HIV-1 exposure. The blots in B and C are representative images from 3 independent experiments, and the quantified results are depicted in bar graphs. Results are means ± SE of 3 experiments. *P < 0.05 or **P < 0.01, data in cultures exposed to HIV-1 are significantly different from the corresponding controls without HIV treatment. †P < 0.05 or ††P < 0.01, data in cultures with inhibited TERT activity by hTERT siRNA or TAG-6 and not exposed to HIV are significantly different from the corresponding control cultures without TERT inhibition. #P < 0.05 or ##P < 0.01, data in cultures with inhibited TERT activity by hTERT siRNA or TAG-6 and treated with HIV-1 are significantly different from the corresponding cultures without TERT inhibition and exposed to HIV-1. ‡‡P < 0.01, data in cultures transfected with hTERT siRNA and treated with HIV-1 in the presence of JSH-23 are significantly different from the corresponding cultures without added JCH-23.

The results of NF-κB transcriptional activity are in agreement with the assessment of p65, the NF-κB binding subunit, in the nuclear extracts of hCMEC/D3 cells exposed to HIV-1 and TAG-6. A combined exposure to HIV-1 and telomerase inhibitor increased p65 levels to higher levels than treatment with HIV-1 or TAG-6 alone (Fig. 4B). The final results of this series of experiments indicate that the inhibition of NF-κB can protect against induction of ICAM-1 by HIV-1 and/or inhibition of TERT activity. As shown in Fig. 4C, treatment with JSH-23, a selective blocker of a nuclear translocation of NF-κB p65, protected against ICAM-1 overexpression induced by hTERT siRNA and exposure to HIV-1-infected U937 cells.

TERT deficiency potentiates HIV-1-induced alterations of tight junction protein expression in brain endothelial cells and brain microvessels.

Tight junctions are the main structural elements that regulate the barrier function of the brain endothelium. A compromised integrity of tight junctions is involved in increased transendothelial migration of inflammatory cells. Therefore, we studied the effects of hTERT silencing and HIV-1-exposure on the expression of tight junction proteins in hCMEC/D3 cells. Exposure to HIV-1-infected U937 cells resulted in a significant decrease in the expression of claudin-5 and ZO-1 in hCMEC/D3 cells (Fig. 5). Importantly, the silencing of hTERT (Fig. 5, A and C) or the preexposure to telomerase inhibitor (Fig. 5, B and D) further potentiated these effects.

Fig. 5.

HIV-induced alterations of tight junction protein expression in hCMEC/D3 cells are potentiated by hTERT inhibition. hCMEC/D3 cells were transfected or treated with TAG-6 as in Fig. 2. The expression of tight junction proteins, claudin-5 (A and B) and zonula occludens-1 (ZO-1; C and D), were determined by Western blot analysis. The blots are representative images from 3 independent experiments, and the quantified results are depicted in the bar graphs. Results are means ± SE of 3 separate experiments. *P < 0.05 or **P < 0.01, data in cultures exposed to HIV-1 are significantly different from the corresponding controls without HIV treatment. †P < 0.05 or ††P < 0.01, data in cultures with inhibited TERT by hTERT siRNA or TAG-6 and not exposed to HIV are significantly different from the corresponding control cultures without TERT inhibition. ##P < 0.01, data in cultures with inhibited TERT by hTERT siRNA or TAG-6 and treated with HIV-1 are significantly different from the corresponding cultures without TERT inhibition and exposed to HIV-1.

The effects of TERT deficiency on the integrity of tight junction proteins were next evaluated in brain capillaries isolated from mice that received injections with HIV-1 protein Tat (or vehicle control) into the ICA. The administration of Tat to wild-type controls resulted in a decrease in immunoreactivity and a fragmentation of staining continuity for claudin-5 and ZO-1. These effects were additionally potentiated in TERT-deficient mice (Fig. 6, A and C, respectively). A quantification by densitometry analyses of Western blots confirmed that Tat-mediated diminished levels of tight junction proteins were statistically significant in TERT-deficient mice compared with wild-type controls (Fig. 6, B and D).

Fig. 6.

HIV Tat-induced disruption of tight junctions is potentiated in brain microvessels of TERT^−/− mice. TERT^−/− and WT mice were injected with HIV-1 Tat or vehicle as in Fig. 3. Brain microvessels were isolated 24 h post-Tat injection and analyzed for claudin-5 and ZO-1 expression using immunofluorescence (A and C) and Western blot analysis (B and D) techniques. The immunostaining data (A and C) are merged immunofluorescence and phase-contrast micrographs acquired using a ×60 oil-immersion lens under identical instrument settings. The blots are representative images from 3 mice per group, and the quantified results are depicted in the bar graphs. Results are means ± SE. **P < 0.01, data in WT or TERT^−/− mice injected Tat are significantly different from those corresponding controls injected with vehicle. ††P < 0.01, data in TERT^−/− mice injected with vehicle are significantly different from the corresponding WT mice injected with vehicle. ##P < 0.01, data in TERT^−/− mice treated with Tat are significantly different from the corresponding WT mice injected with Tat.

MMP-9 is involved in claudin-5 disruption by HIV-1 and telomerase inhibition.

Increased MMP-9 activity has been linked to HIV-1-induced alterations of the integrity of the brain endothelium (15, 18, 25). Therefore, we evaluated the effects of telomerase inhibition on MMP-9 activity and the role of these effects in the disruption of tight junction proteins. The exposure of hCMEC/D3 cells to HIV-1 or the telomerase inhibitor TAG-6 increased MMP-9 promoter activity in cells transfected with the pGL3 MMP-9 construct. In addition, the inhibition of telomerase activity potentiated HIV-1-induced stimulation of the MMP-9 promoter (Fig. 7A). To evaluate the role of NF-κB in these events, selected hCMEC/D3 cultures were transfected with the MMP-9 promoter construct with mutated NF-κB binding sites (pGL3 mt-MMP-9). Such a mutation completely blocked the upregulation of MMP-9 induced by HIV-1 and/or telomerase inhibition (Fig. 7A).

Fig. 7.

Matrix metalloproteinase-9 (MMP-9) is involved in disruption of claudin-5 expression by HIV-1 and telomerase inhibition. A: hCMEC/D3 cells were transfected with MMP-9 promoter construct (pGL3 MMP-9) or pGL3 MMP-9 with mutated NF-κB binding sites (pGL3 mt-MMP-9). In addition, selected cultures were pretreated with TAG-6 (2 μM) for 2 h and exposed to HIV-1 particles (200 pg p24/ml) for 24 h. B: hCMEC/D3 cells were transfected with control or hTERT siRNA and exposed to HIV-1 as in Fig. 1. MMP-9 activity was assessed by zymography. C: hCMEC/D3 cells were treated as in B. In addition, selected cultures were pretreated for 2 h with GM6001 (general inhibitor of MMP activity; 5 μM) or iMMP-9 (a selective inhibitor of MMP-9, 5 μM). Claudin-5 expression was then determined by Western blot analysis. The blots in B and C are representative images from 3 independent experiments, and the quantified results (means ± SE) are depicted in the bar graphs. *P < 0.05, **P < 0.01, or ***P < 0.001, data in cultures exposed to HIV-1 are significantly different from the corresponding controls without HIV treatment. †P < 0.05, data in cultures with inhibited TERT by hTERT siRNA or TAG-6 and not exposed to HIV are different from the corresponding control cultures without TERT inhibition. ##P < 0.01, data in cultures with inhibited TERT by hTERT siRNA or TAG-6 and treated with HIV-1 are different from the corresponding cultures without TERT inhibition and exposed to HIV-1. ‡‡P < 0.01, data in cultures transfected with hTERT siRNA and treated with HIV-1 in the presence of iMMP-9 or GM6001 are different from the corresponding cultures without added inhibitors.

The role of telomerase inhibition in MMP-9 activity was also evaluated using zymography. hCMEC/D3 cells were transfected with control or hTERT siRNA and coexposed to HIV-1, MMP-9-specific inhibitor (iMMP-9; 5 μM; Calbiochem, San Diego, CA), and/or a general MMP inhibitor (GM6001; 5 μM; Biomol, Plymouth Meeting, PA). As shown in Fig. 7B, the exposure to HIV-1 induced MMP-9 activity that was significantly enhanced by hTERT silencing. Most importantly, the inhibition of MMP-9 activity significantly protected against the downregulation of claudin-5 levels in hCMEC/D3 cells transfected with hTERT siRNA and exposed to HIV-1 (Fig. 7C).

DISCUSSION

In the present study, we hypothesized that endothelial senescence induced by the inhibition of telomerase activity may contribute to the transendothelial migration of HIV-1 via increased proinflammatory responses. Normal endothelial cells, as most somatic cells, do not express telomerase activity. However, the hCMEC/D3 cell line was generated by incorporating hTERT by lentiviral transduction. Thus the modulation of telomerase activity in these cells either by hTERT silencing or the use of pharmacological inhibition provides a model of endothelial cell senescence associated with the loss of telomerase activity and the shortening of telomeres.

The induction of inflammatory reactions and the disruption of BBB integrity are the critical events in HIV trafficking into the brain (18, 43, 53). The results from our (3, 26) and other laboratories (46, 47) indicate that both HIV-1 and HIV-1-specific protein Tat can induce proinflammatory reactions in brain endothelial cells. Among inflammatory mediators, ICAM-1 plays a critical role in the firm attachment and transendothelial migration of leukocytes (1, 43). Our data indicate that TERT deficiency can contribute to an HIV-1-induced upregulation of ICAM-1 in both brain endothelial cells and brain microvessels (Figs. 2 and 3). These effects are in agreement with literature reports demonstrating the importance of telomere function in the vasculature. For example, it was shown that the senescence of endothelial cells resulted in an elevated expression of ICAM-1 and diminished endothelial nitric oxide synthase activity (32). Nevertheless, it should be pointed out that hCMEC/D3 cells were exposed to HIV-1-infected U937 cells in the majority of the in vitro experiments in the present study. Under these experimental conditions, it is difficult to fully distinguish whether the observed effects were induced solely by HIV-1 treatment or by inflammatory factors released from infected U937 cells.

hTERT silencing resulted in an HIV-1-mediated upregulation of ICAM-1 at both mRNA and protein levels. This transcriptional upregulation prompted us to evaluate the activation of NF-κB, the principal transcription factor that regulates the expression of several inflammatory mediators. Novel results in our study indicate that an HIV-1-induced transactivation of NF-κB responsive construct is potentiated in cells with silenced hTERT (Fig. 4A). NF-κB can mediate telomerase activity via the activation of SP-1 and c-Myc transcription factors (42). Thus it appears that the activation of NF-κB may be responsible for the dramatic upregulation of inflammatory responses and the transendothelial migration of U937 cells in hCMEC/D3 cells with inhibited hTERT and exposed to HIV-1.

The integrity of tight junction proteins is another critical factor that regulates HIV-1 trafficking into the brain (16, 25, 29, 36, 45, 56). We recently demonstrated that the exposure of hCMEC/D3 cells to HIV-infected monocytes resulted in a decreased expression of tight junction proteins, such as junctional adhesion molecule-A, occludin, and ZO-1 (25). In the present study, we observed that an HIV-1-induced disruption of tight junction protein expression was markedly potentiated by the inhibition or deficiency of TERT (Fig. 6). Interestingly, TERT deficiency affected the integrity of both claudin-5, i.e., the transmembrane tight junction protein that forms the backbone of tight junctions (28), and ZO-1 that links the transmembrane tight junction proteins to the cytoskeleton. These results suggest that endothelial senescence contributes to the dysfunction of the BBB, especially when an additional factor, such as HIV-1, is present. Indeed, a region-specific disruption of the BBB integrity was observed in a mouse model of senescence (4). In addition, functional alterations of the BBB were observed in healthy older volunteers compared with young subjects (5). Aging was also shown to be associated with the deterioration of the blood-retinal barrier function, as evidenced by leakage of the intravascular tracer into the retinal parenchyma and reduced immunoreactivity for occludin in rats (11).

To address the mechanisms of decreased tight junction protein expression due to HIV-1 treatment and telomerase inhibition, our studies focused on the regulatory role of MMP-9. MMP-9 can degrade extracellular matrix components, facilitate the migration of inflammatory cells across the endothelium, and contribute to HIV-1 pathology (38). Elevated levels of MMP-9 have been reported in the cerebrospinal fluid of HIV-1-infected children (31) and adult patients (44). In addition, increased brain levels of MMP-9 are more frequent in patients with neurological complications of HIV-1 infections, such as HIV-associated dementia (15). Consistent with these reports, the exposure to HIV-1 increased the MMP-9 promoter and enzyme activities in the present study (Fig. 7). Importantly, these effects were markedly potentiated by telomerase inhibition or silencing. MMP expression is controlled primarily at the transcriptional level, with NF-κB being one of the main transcriptional factors involved in such a regulation (33, 49). Indeed, the mutation of the NF-κB binding sites (at −610 and −340) in the MMP-9 promoter protected against the upregulation of MMP-9 promoter activity by HIV-1 and telomerase inhibition (Fig. 7).

Recent evidence, partially generated in our laboratory, implicated individual MMPs in the regulation of tight junction expression (23, 25, 55). The increased MMP expression was shown to lead to the degradation of tight junction proteins and the opening of the BBB in a hypoxia/reperfusion model of stroke (21, 55). Moreover, oxidative stress can activate selected MMPs, enhance tyrosine phosphorylation of tight junction proteins, and thus disrupt the BBB (23). The results of the present study strongly indicate that increased MMP-9 activity is involved in the disruption of tight junction proteins in hCMEC/D3 cells with silenced TERT and exposed to HIV-1. In fact, both a broad-spectrum inhibitor GM6001 and a specific anti-MMP-9 inhibitor protected against HIV-induced alterations of claudin-5 levels (Fig. 7C).

In conclusion, the senescence of brain endothelial cells induced by the inhibition of telomerase activity in hCMEC/D3 cells significantly increased HIV-1-stimulated transendothelial migration of monocytes. These effects were associated with the upregulation of ICAM-1 and MMP-9 activity that resulted in a downregulation of tight junction protein expression. It appears that the activation of the transcription factor NF-kB by HIV-1 and telomerase inhibition may be the underlying mechanism leading to the induction of inflammatory responses and the upregulation of MMP-9. The present data indicate that senescence of brain endothelial cells may contribute to HIV-1-induced dysfunction of the BBB.

DISCLOSURES

No conflicts of interest are declared by the author(s).