Suppression of Transforming Growth Factor β Receptor 2 and Smad5 Is Associated with High Levels of MicroRNA miR-155 in the Oral Mucosa during Chronic Simian Immunodeficiency Virus Infection

Abstract

Chronic human immunodeficiency virus and simian immunodeficiency virus (HIV and SIV) infections are characterized by mucosal inflammation in the presence of anti-inflammatory cytokines such as transforming growth factor β (TGFβ). The mechanisms for refractiveness to TGFβ are not clear. Here we show that the expression of microRNA miR-155 was significantly upregulated in the oropharyngeal mucosa during chronic SIV infection and was coincident with downregulation of TGFβ receptor 2 (TGFβ-R2) and SMAD5, key TGFβ signaling genes that harbor putative target sites for miR-155. Ectopic expression of miR-155 in vitro was found to significantly downregulate TGFβ-R2 and Smad5 expression, suggesting a role for miR-155 in the suppression of TGFβ-R2 and SMAD5 genes in vivo. The downregulation of TGFβ signaling genes by miR-155 likely contributes to the nonresponsiveness to TGFβ during SIV infection and may inadvertently aid in increased immune activation during HIV and SIV infections.

TEXT

Chronic human immunodeficiency virus (HIV) infection is characterized by a significant increase in generalized immune activation (1, 2) and in mucosal inflammatory conditions such as periodontitis and gingivitis (3–6). Some studies have reported an increased migration of inflammatory cells into the oral mucosa (7), whereas others have shown that levels of proinflammatory cytokines such as interleukin-6 (IL-6) (8–10) were significantly elevated during HIV infection. Interestingly, chronic immune activation and mucosal inflammation occurs in the presence of transforming growth factor β (TGFβ), a highly immunosuppressive cytokine (11, 12) that has been shown to deactivate mononuclear phagocytes (13), reduce IL-2 receptor expression on T cells, and inhibit IL-2-induced T cell proliferation (14, 15).

Numerous studies have shown that TGFβ levels were elevated during HIV infection and were associated with disease progression (16–18). Titers of plasma TGFβ were shown to be sufficient to induce cellular responses in vitro (16). Likewise, simian immunodeficiency virus (SIV)-infected rhesus macaques were found to have persistently increased levels of TGFβ in the plasma during chronic infection (19).

Why immune activation and chronic inflammation persist even in the presence of high levels of immunosuppressive cytokines is not clear. Ploquin et al. (20) reported that the failure of TGFβ to resorb virus-driven inflammation and activation during pathogenic HIV type 1 (HIV-1) and SIV infections may be due to increased unresponsiveness to TGFβ that was associated with altered signaling. The exact mechanisms for this unresponsiveness are still not clear.

TGFβ binds to and signals through its high-affinity type I and type II serine/threonine kinase receptors TGFβ receptor 1 (TGFβ-R1) and TGFβ-R2, which are expressed on a variety of cell types (21, 22). Binding of TGFβ to TGFβ-R2 phosphorylates and activates TGFβ-R1, which in turn recruits and phosphorylates a number of receptor-activated Smad proteins such as Smad1 to Smad5 (23). We hypothesized that the inability of TGFβ to suppress immune activation may be associated with the dysregulated expression of its receptors in the oral mucosa. The loss of TGFβ receptors has been shown to mediate resistance to TGFβ in cell lines (22, 24). We used the rhesus macaque model of HIV infection to address this question.

Seventeen rhesus macaques (Macaca mulatta) of Indian origin (n = 17) were used in this study. Ten animals were chronically infected with pathogenic SIVmac251, whereas the rest (n = 7) were uninfected. The animals were housed in accordance with the guidelines of the American Association for Accreditation of Laboratory Animal and were seronegative for SIV, simian retrovirus (SRV), and simian T-cell leukemia virus (STLV) type 1 prior to SIV challenge.

Peripheral blood and oropharyngeal tissue samples were collected at the time of sacrifice. Plasma and peripheral blood mononuclear cells (PBMC) were isolated and cryopreserved along with the oropharyngeal tissue samples. PBMC were isolated by density gradient centrifugation, whereas cells from the oropharyngeal mucosa were isolated by enzymatic digestion and Percoll gradient centrifugation per procedures described previously (25–30).

IL-6 and TGFβ levels are upregulated but TGFβ-R2 and Smad5 expression levels are downregulated in the oropharyngeal mucosa during SIV infection.

SIV-infected animals had ∼6 logs of SIV RNA copies/ml of plasma (Fig. 1a). Plasma viral loads were determined by real-time PCR using reverse-transcribed (RT) viral RNA as the template, as previously described (31). The ratios of CD4/CD8 T cells in peripheral blood and the oropharyngeal mucosa were determined by flow cytometry (Fig. 1b and c). Briefly, isolated cells were labeled with specific panels of antibodies containing a combination of CD45-PE (phycoerythrin), CD3-Cy7APC (allophycocyanin), CD4-PB, CD8-Alexafluor 700 or CD45-PE, CD3-Cy7APC, CD20-APC, and Ki-67-FITC (fluorescein isothiocyanate). All the antibodies were obtained from BD Biosciences (San Diego, CA) and were titrated using rhesus macaque PBMC. Flow cytometric data were analyzed using FlowJo version 9.2 (Tree Star, Inc., Ashland, OR). Statistical analysis was performed using the Mann-Whitney U test with GraphPad Prism Version 4.0 software (GraphPad Prism Software, Inc. San Diego, CA). We were unable to obtain absolute CD4 T cell counts due to lack of cumulative blood count (CBC) data. The frequencies of CD4 T cells in both peripheral blood and oropharyngeal mucosa were significantly reduced during chronic SIV infection (Fig. 1b and c).

FIG 1.

Expression of IL-6 mRNA is elevated in the oral mucosa during chronic SIV infection. (a) Plasma viral loads (the limit of detection is 30 copies/ml of plasma). (b and c) The ratios of CD4/CD8 T cells in the oropharyngeal mucosa (b) and peripheral blood (c) at the time of sacrifice. (d) IL-6 mRNA expression in the oropharyngeal mucosa.

We next examined if the levels of IL-6, a proinflammatory cytokine, were altered in the oropharyngeal mucosal tissue of SIV-infected animals compared to uninfected animals as a number of studies had shown elevated levels of IL-6 during infection (8–10). IL-6 mRNA expression was determined by a relative quantitative reverse transcription-PCR (qRT-PCR) assay. RNA was isolated from oropharyngeal tissue using an miRNeasy kit (Qiagen Sciences, Gaithersburg, MD) and treated with Ambion Turbo DNase (Applied Biosystems, Austin, TX) to remove contaminating DNA. Each DNase-treated RNA sample was tested in a qRT-PCR assay using β-actin primers and a probe to confirm that RNA was free from DNA contamination. Purified RNA was reverse transcribed using a Superscript III First Strand Synthesis kit (Invitrogen, Carlsbad, CA) and used in a TaqMan relative qPCR assay with high-fidelity Platinum Taq polymerase (Invitrogen) on an ABI 7500 instrument (Applied Biosystems) as described previously. Macaca mulatta-specific primers and probes (Table 1) were used to determine gene expression, with the β-actin housekeeping gene used as controls. All primers and probes were designed using Primer-3 software (32). Collected data were analyzed using the 2^−ΔΔCT (ddCT) threshold cycle method, and fold change in expression was calculated as previously described (33).

TABLE 1.

Rhesus macaque-specific primer and probe sequences

Gene product	Forward primer (5′ to 3′)	Reverse primer (5′ to 3′)	FAM-probe-BHQ1 (5′ to 3′)a
rh β-actin	ATGCTTCTAGGCGGACTGTG	AAAGCCATGCCAATCTCATC	TGCGTTACACCCTTTCTTGACAAAACC
rh IL-6	ATGCAATAACCACCCCTGAA	AAGAGCCCTCAGGTTGGACT	TGCTGACGAAGCTGCAGGCA
rh TGFβ	TGTCATAGATTTCGTTGTGGGTTT	GTACAACAGCACCCGCGAC	ACCATTAGCACGCGGGTGACCTCC
rh TGFβ-R1	AAGGCCAAATATCCCAAACA	TAGCTGCTCCGTTGGCATAC	AAGGCTTCGCAGCTCTGCCA
rh TGFβ-R2	ATAGGACTGCCCATCCACTG	CAGGCAGGATTTCTGGTTGT	TGGTCACTGACAACAATGGTGCA
rh Smad5	ATTGTTGGGCTGGAAACAAG	TTCACCAAAGCATCAACTGC	TGATGAGGAGGAGAAATGGGCAGA

^aFAM, 6-carboxyfluorescein; BHQ1, black hole quencher 1; rh, rhesus macaque.

We observed a significant increase in the expression of IL-6 in the oropharyngeal mucosa of SIV-infected animals compared to uninfected animals (Fig. 1d). To determine if high levels of IL-6 expression were associated with altered levels of TGFβ expression, we examined TGFβ levels in the oropharyngeal mucosa of SIV-infected animals and compared them to the TGFβ levels in the oropharyngeal mucosa of SIV-uninfected animals. SIV-infected animals were found to express ∼40× more TGFβ transcripts in the oropharyngeal mucosa than SIV-uninfected animals (Fig. 2a).

FIG 2.

TGFβ levels are upregulated but TGFβ-R1, TGFβ-R2, and Smad5 expression is downregulated in the oropharyngeal mucosa of SIV-infected animals. (a) TGFβ mRNA levels in the oropharyngeal mucosa of SIV-infected animals relative to SIV-negative animals. (b and c) Representative dot plots showing the gating strategy used to examine Ki-67 expression on T cells (b) and frequency of CD45⁺ CD3⁺ Ki-67 T cells in the oropharyngeal mucosa (c). (d) Expression of TGFβ-R1, TGFβ-R2, and Smad5 mRNA in the oropharyngeal mucosa of SIV-infected animals relative to SIV-negative animals.

TGFβ has been shown to have antiproliferative effects on T cells (14, 15). To determine if high levels of TGFβ expression in the oral mucosa were associated with suppression of proliferative responses, we examined the expression of Ki-67 on CD45⁺ CD3⁺ T cells in SIV-infected animals and compared the expression to that seen in healthy animals (Fig. 2b and c). Ki-67 is a nuclear antigen expressed by proliferating T cells and has been extensively used as a marker of immune activation during HIV and SIV infections. Ki-67 expression was determined by intracellular staining using a Cytofix/perm kit from BD Biosciences. Labeled cells were fixed with 0.5% paraformaldehyde and analyzed using a Becton Dickinson LSR II flow cytometer. Our results showed that levels of CD45⁺ CD3⁺ Ki-67⁺ T cells were significantly increased in the oral mucosa, suggesting that the immune activation observed in other mucosal sites is apparent in the oral mucosa during chronic SIV infection.

As immune activation persisted in the presence of high levels of TGFβ, we hypothesized that TGFβ signaling may have been dysregulated in the oral mucosa during chronic SIV infection. As TGFβ signals through its receptors, we examined the expression levels of TGFβ-R1 and -R2 mRNA in the oropharyngeal mucosa of SIV-infected animals and compared them to those of SIV-uninfected animals (Fig. 2d). We observed a significant decrease in the expression levels of both TGFβ receptors in SIV-infected animals, suggesting that altered signaling may be a potential mechanism mediating the refractiveness to TGFβ. We next examined the mRNA levels of Smad5, a receptor-activated Smad protein that plays a role in TGFβ signaling, to determine if SIV infection was associated with suppression of Smad expression in the oral mucosa. Our results showed that expression of Smad5 mRNA was significantly suppressed in the oral mucosa of chronically infected animals (Fig. 2d) relative to healthy animals.

MicroRNA miR-155 expression is upregulated in the oropharyngeal mucosa during SIV infection.

Previous studies have shown that TGFβ receptors and Smad5 were posttranscriptionally regulated by microRNAs such as miR-155 (34, 35). MicroRNAs are small noncoding RNAs that posttranscriptionally regulate the expression of cellular genes. A number of these small RNAs have been implicated in mediating inflammation (36, 37), and miR-155 levels were shown to be upregulated during inflammatory responses (38, 39). Billeter et al. (40) showed that miR-155 played a role in potentiating inflammatory responses. Other studies have shown that miR-155 expression is induced in response to the presence of inflammatory stimuli or activation of Toll-like receptors (TLR) or cytokine stimulation (38, 41) and plays a role in regulating germinal center responses (42, 43).

To determine if miR-155 levels in the oropharyngeal mucosa were altered during chronic SIV infection, we examined the levels of miR-155 in SIV-infected animals and compared them to those in uninfected animals (Fig. 3a). Purified RNA was isolated as described above and reverse transcribed using a miScript II reverse transcription kit (Qiagen) per the instructions of the manufacturer. Synthesized cDNA was used in an Macaca mulatta l_miR-155_1 (Mml_miR-155_1) miScript primer assay (Qiagen) with M. mulatta-specific miR-155 primers (5′-UUAAUGCUAAUCGUGAUAGGGGU-3′) and a miScript SYBR green PCR kit (Qiagen). Non-species-specific RNU1a1 small nuclear RNA (Qiagen) was used as an endogenous control. A real-time qRT-PCR assay was carried out at 95°C for 5 min, followed by 40 cycles of 95°C for 15 s and 60°C for 30 s, on an ABI 7500 instrument (Applied Biosystems). Collected data were analyzed using the 2^−ΔΔCT (ddCT) method, and fold change in expression was calculated as previously described (33). Our results showed that SIV-infected animals had significantly higher levels of miR-155 expression in the oropharyngeal mucosa than SIV-uninfected animals (Fig. 3a). Bignami et al. (44) showed that HIV-infected subjects have high levels of cellular miR-155 compared to multiply exposed but uninfected individuals. On the other hand, Seddiki et al. (45) reported that HIV-infected patients display significantly elevated levels of T regulatory cells with upregulated expression of miR-155 that was associated with disease pathogenesis.

FIG 3.

miR-155 expression is upregulated in the oropharyngeal mucosa of SIV-infected animals. (a) miR-155 expression in the oropharyngeal mucosa of SIV-infected animals relative to SIV-negative animals. (b) Alignment showing miR-155 seed sequence target sites on rhesus macaque TGFβ-R2 and Smad5 3′UTR. (c and d) Expression of miR-155 (c) and TGFβ-R1, TGFβ-R2, Smad5, and CD4 (d) mRNA in primary PBMC transfected with pmiR-155 plasmid relative to controls.

We next sought to determine if the elevated levels of miR-155 were the likely cause for the suppression of TGFβ receptors and Smad5 expression. To address this question, we first examined whether rhesus macaque TGFβ-R1, TGFβ-R2, and Smad5 harbored target sites in the 3′ untranslated region (3′UTR) for miR-155 using Targetscan and miRBASE. In line with studies in humans, only rhesus macaque TGFβ-R2 and Smad5 3′UTR and not TGFβ-R1 were found to have a target site for the seed sequence of miR-155 (Fig. 3b).

To determine if miR-155 could suppress the expression of TGFβ-R2 and Smad5, we ectopically expressed miR-155 in phytohemagglutinin (PHA)-activated primary rhesus macaque PBMC by transfection with pCDNA3.1-miR-155 expression plasmids (pmiR-155). PBMC transfected with the empty vector were used as controls. Ectopic expression of miR-155 was confirmed by qRT-PCR assay. Briefly, PBMC from 5 healthy animals were cultured in 10% Dulbecco's modified Eagle's medium (DMEM) in 12-well culture plates in a volume of 1 ml at a density of 5 × 10⁵ cells per well in the presence of PHA (10 μg/ml) for 48 h. Cells were harvested after culture, and transfections were performed with FuGENE HD transfection reagent according to the manufacturer's instructions with modifications. Approximately 6 μl of FuGENE HD Transfection Reagent was added to 500 μl serum-free DMEM, mixed, and incubated at room temperature for 5 min. Transfected cell cultures were set up in the 500-μl transfection mix as follows: (i) 2 μg of plasmid vector (800 ng pcDNA 3.1 miR-155 plasmid plus 1,200 ng of control plasmid); (ii) 2 μg of pcDNA 3.1 control plasmid; (iii) untransfected cells in 500 μl of serum-free DMEM; and (iv) cells in transfection mix only. Transfected cells were incubated at 37°C for 5 days under 5% CO₂. Cells were harvested at the end of culture and used for determining miR-155 and cellular gene expression as described above.

Both miR-155 expression and sensor plasmids that included oligonucleotide primers were generated as previously described (46). For the construction of pmiR-155 expression vector, ∼180 nucleotides (nt) encompassing the stem-loop pre-miRNA were PCR amplified from BCBL-1 genomic DNA. PCR products were TOPO cloned, excised with HindIII and XhoI, and inserted into pcDNA3.1V5/HisA (Invitrogen) at the corresponding sites. The miR-155 sensor plasmids were constructed using pGL3-Promoter (Promega). Oligonucleotides containing two complementary miR-155 sites were annealed and inserted at FseI and XbaI sites downstream of the luciferase gene.

Rhesus PBMC transfected with pmiR-155 was found to express ∼8× more miR-155 than PBMC transfected with the empty control plasmid (Fig. 3c). We next examined the expression levels of TGFβ-R1, TGFβ-R2, and Smad5 in the same pmiR-155-transfected PBMC. The CD4 gene, which does not have a target site for miR-155 in its 3′ UTR, was used as a control. Ectopic expression of miR-155 was accompanied by a significant suppression of both TGFβ-R2 expression and Smad5 expression with no major effect on the level of either TGFβ-R1 or CD4 mRNA expression (Fig. 3d).

It is not clear from our studies why TGFβ-R1 was suppressed in vivo, as TGFβ-R1 was not found to harbor a target site for miR-155 in its 3′ UTR. TGFβ-R1 is recruited after activation of TGFβ-R2, and it is possible that the suppression of TGFβ-R2 is accompanied by lower TGFβ-R1 expression. Likewise, Smad5 is phosphorylated and activated by TGFβ-R1 and it is possible that the suppression of TGFβ-R1 contributed to the lower levels of Samd5 expression. Though the exact mechanisms for suppressed expression of TGFβ-R1 are not clear from our studies, our results suggest that multiple mechanisms, including regulation by miR-155, may contribute to the altered TGFβ signaling during chronic SIV infection.

Taken together, our studies suggest that chronic SIV infection is characterized by nonresponsiveness to immunosuppressive cytokines such as TGFβ and higher levels of proinflammatory cytokines such as IL-6 that contribute to the overall inflammation of the oral mucosa. We identify a potential role for miR-155-mediated suppression of TGFβ-R2 and key signaling molecules such as Smad5 in this process.