Combination antiretroviral therapy (cART) restores HIV-1 infection-mediated impairment of JAK-STAT signaling pathway

JAK-STAT signaling pathway has a crucial role in host innate immunity against viral infections, including HIV-1. We therefore examined the impact of HIV-1 infection and combination antiretroviral therapy (cART) on JAK-STAT signaling pathway. Compared to age-matched healthy donors (n = 18), HIV-1-infected subjects (n = 18) prior to cART had significantly lower expression of toll-like receptors (TLR-1/4/6/7/8/9), the IFN regulatory factors (IRF-3/7/9), and the antiviral factors (OAS-1, MxA, A3G, PKR, and Tetherin). Three months’ cART partially restores the impaired functions of JAK-STAT-mediated antiviral immunity. We also found most factors had significantly positive correlations (p < 0.05) between each two factors in JAK-STAT pathway in healthy donors (98.25%, 168/171), but such significant positive associations were only found in small part of HIV-1-infected subjects (43.86%, 75/171), and stably increased during the cART (57.31%, 98/171 after 6 months’ cART). With regard to the restoration of some HIV-1 restriction factors, HIV-1-infected subjects who had CD4+ T cell counts > 350//μl responded better to cART than those with the counts < 350/μl. These findings indicate that the impairment of JAK-STAT pathway may play a role in the immunopathogenesis of HIV-1 disease.


INTRODUCTION
Innate immunity plays a critical role in host defense mechanisms against viral infections, including HIV-1 [1]. The innate immune system includes three classes of receptors, Toll-like receptors (TLRs), retinoic acidinducible gene I (RIG-I)-like receptors (RLRs), and nucleotide oligomerization domain (NOD)-like receptors (NLRs) [2]. Among these receptors, TLRs function as the primary sensors of pathogens, activation of which can stimulate several signaling pathways [3], including the Janus kinase-signal transducer and activator of transcription (JAK-STAT).
It has been documented that HIV-1 infection could immediately trigger strong innate and adaptive immune responses [21]. In vitro studies revealed that IFN-λ can upregulate cellular anti-HIV-1 factors through JAK-STAT signaling pathway and significantly inhibited HIV-1 infection [1,12]; TLR-3 activated by Poly I:C can induce multiple anti-HIV-1 factors and inhibit HIV-1 infection of macrophages [22]. These in vitro findings indicate that host innate immunity is critical in restricting HIV-1 replication and spread. However, by interfering with the signaling moleculars through its proteins, HIV-1 can suppress the induction of IFNs and antiviral ISGs and persist in immune cells. Our previous in vitro study [12] demonstrated that JAK-STAT pathway is involved in the induction of the anti-HIV-1 cellular factors and HIV-1 inhibition in macrophages [23]. Thus, it is of importance to determine whether HIV-1 infection impairs the JAK-STAT signaling pathway and whether cART can reverse HIV-1 infection-mediated injury of JAK-STAT pathway.

Subject information
Eighteen HIV-1-infected subjects had CD4+ T cell counts of 62~670 /μL (mean of 362) and viral loads of 0.081~79 × 10e4 copies/mL (mean of 6.9 × 10e4) at the time of study enrollment (Table 1). They also had normal aspartate transaminase (AST, 27.4 ± 18.7 IU/mL) and alanine transaminase (ALT, 26.6 ± 16.1 IU/mL), and were satisfied for antiviral treatment. The majority (17 out of 18) of HIV-1-infected subjects were men who have sex with men (MSM) and treated with TDF+3TC+EFV. During the course of study, neither liver injury (AST < 72 IU/L) nor clinical symptoms were reported in these study subjects. Agematched healthy donors (n = 18) were enrolled as the control subjects.

Effect of HIV-1 infection or cART on TLRs and IRFs
TLRs can recognize viral RNA or DNA and initiate antiviral signaling pathways [7,12,24]. We thus examined six TLRs (TLR-1/4/6/7/8/9) that are known to be involved in antiviral immunity [9,24,25]. Figure 1 showed that peripheral blood mononuclear cells (PBMCs) from HIV-1-infected subjects before or one month after cART had lower levels of the TLRs than those of the control subjects. Three-month cART could reverse the levels of the TLRs. However, these levels of TLRs were still lower than those of uninfected subjects. We next analyzed the expression of three IFN regulatory factors (IRF-3, IRF-7 and IRF-9) that have critical role in the regulation of IFN. As shown in Figure 2, HIV-1-infected subjects had significant lower levels of IRF-7 and IRF-9 than the control subjects. Three-month treatment with cART could restore the levels of these IRFs ( Figure 2).

Impact of HIV-1 infection or cART on anti-HIV-1 factors
Several cellular factors (OAS-1, MxA, APOBEC3G (A3G), Tetherin, protein kinase (PKR), and interferon-stimulated gene 56 (ISG56)) in JAK-STAT pathway have been shown to have anti-HIV-1 activities. We thus examined the expression of these factors in PBMCs of the study subjects. As shown in Figure 3, HIV-1-infected subjects had significantly lower levels of the HIV-1 restriction factors than the control subjects. Three-month cART could significantly increase the levels of these factors in PBMCs of HIV-1-infected subjects ( Figure 3). Comparing with HIV-1-infected subjects at prior to or after cART (1, 3 or 6 months), uninfected subjects had the highest levels of the HIV-1 restriction factors ( Figure 3).

The association to CD4+ T cells counts or HIV-1 loads
To explore the association of JAK-STAT pathway with disease process, we further examined the correlation of cellular factors in JAK-STAT pathway with CD4+ T cells and plasma viral loads. CD4+ T cells counts were increased to 444 ± 192 /μL after three-month cART and to 460 ± 199/μL after six-month cART as compared to 362 ± 163/μL prior to cART (Table 1). In contrast, HIV-1 loads were decreased to 36 ± 57 copies/mL 3 months after cART and to 28 ± 70 copies/mL after six-month cART, as compared to 69184 ± 184778 copies/mL prior to cART (Table 1). Although no significant correlation was found between antiviral factors and CD4+ T cell counts or HIV-1 loads in the study subjects prior to or after cART, the subjects with > 350/μl CD4+ T cells counts showed better response (p < 0.05) to cART after three-months' treatment ( Figure 5). Specifically, after three-months' antiviral therapy, the HIV-1-infected subjects with high CD4+ T cells counts (CD4 > 350/μl, n = 10) prior to cART had higher expression of cellular factors in JAK-STAT pathway than that of subjects with low CD4 + T cells count (CD4 < 350/μl, n = 8), especially for IRF-3/7/9, TLR-1/4/6/8, OAS-1, MxA, PKR, Tetherin, and STAT-2 (p < 0.05) ( Figure 5).

DISCUSSION
In the present study, we examined the in vivo impact of HIV-1 infection on JAK-STAT signaling pathwaymediated anti-HIV-1 immunity. JAK-STAT signaling pathway has been implicated in TLR-mediated pathogen recognition, IRFs activation [26], the induction of IFNinducible genes such as MxA and ISG56 [12,22]. We found that HIV-1-infected subjects had lower levels of the TLRs, IRFs and the cellular anti-HIV factors (Figures 1, 2, 3). These findings indicate that HIV-1infected subjects had a compromised JAK-STAT-mediated antiviral immunity. Interestingly, cART could only restore some of the HIV-1 restriction factors (Figure 3). We further analyzed the correlations between each two factors in JAK-STAT pathway, including adaptor molecular such as MyD88 and STAT-2. As shown in Figure 4 and Supplementary Table 1, significantly positive correlations were found in most factors in the control subjects, but only less than half of factors in HIV-1-infected subjects, and stably increased with the cART. These data were in agreement with the expression of TLRs, IRFs, and anti-HIV-1 factors in JAK-STAT pathway, indicating that cART can restore the impaired JAK-STAT pathway in PBMCs of HIV-1-infected subjects. While the mechanism(s)

Combination antiretroviral therapy (cART)
TDF+3TC+EFV 17 -AZT+3TC+NVF 1 -Data was expressed as mean ± SD. * There was no age difference between two groups (p = 0.1236). www.impactjournals.com/oncotarget The expression levels were shown relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and the significant differences from paired or unpaired t tests were shown by *p < 0.05, **p < 0.01, and ***p < 0.001. www.impactjournals.com/oncotarget  18) were set as a control group. Cellular RNAs extracted from PBMCs were subjected to quantitative PCR for the indicated IRFs. The expression levels were shown relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and the significant differences from paired or unpaired t tests were shown by *p < 0.05, **p < 0.01, and ***p < 0.001. involved in HIV-1 infection-mediated impairment of JAK-STAT pathway remain to be determined, it has been shown that Vpr and Vif, two HIV-1 accessory proteins could bind to TANK-binding kinase 1 (TBK1) and inhibit TBK1 autophosphorylation [27], which is a key step in the activation of IFN and NF-κB signaling pathways. Also, HIV-1 Vpu accessory protein is known to induce IRF3 cleavage [28]. The HIV-1-infected subjects with high CD4+ T cells count (> 350/μl) appeared to respond better to cART in terms of the restoration of some of the antiviral factors in JAK-STAT pathway. These findings are clinically significant as they indicate that cART is beneficial for HIV-1-infected subjects who have CD4+ T cell counts > 350/μl than those with the counts < 350/μl. However, further studies with more clinical specimens are needed in order to determine clinical beneficial of early cART in the restriction of host innate immunity against HIV-1.

Figure 3: Cellular anti-HIV-1 factors expression in PBMCs of HIV-1-infected subjects on cART. PBMCs were collected
from HIV-1-infected subjects (n = 18) prior to (0 m) or 1 month (1 m), 3 months (3 m), and 6 months (6 m) after cART. Age-matched healthy subjects (n = 18) were used as a control group. Cellular RNAs extracted from PBMCs were subjected to quantitative PCR for anti-HIV-1 factors (OAS-1, MxA, A3G, Tetherin, PKR, ISG56). The expression levels were shown relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and the significant differences from paired or unpaired t tests were shown by *p < 0.05, **p < 0.01, and ***p < 0.001.  After three months' cART, cellular RNAs extracted from PBMCs were subjected to quantitative PCR for the factors in JAK-STAT pathway. Error bars represent SD, with significance from t test shown by *p < 0.05 and **p < 0.01.

Subjects
Eighteen HIV-1-infected subjects (17 males and one female) and 18 age-matched healthy donors (16 males and two females) were enrolled if they aged 18 or over and volunteered with written informed consent. The study was approved by the Ethics Committee of Wuhan Centers for Disease Prevention & Control. All HIV-1-infected subjects were examined by medical screening, and voluntarily received cART. The treatment regimen and algorithm TDF+3TC+EFV (17 patients) or AZT+3TC+NVP (1 patient) were carried out according to the China free cART manual [29], in which the threshold for cART eligibility was recommend as CD4+ T cells < 350/μL [30] and suggested therapy regimen was AZT/d4T+3TC+NVP/EFV.

Blood sampling
For HIV-1-infected subjects, whole blood was collected prior to cART (0 month), 1 month, 3 months, and 6 months after cART. Estimation of absolute CD4+ T cell counts and plasma HIV-1 loads were tested as described previously [31]. Whole blood (10 mL) of HIV-1-infected subjects and healthy donors were subjected for the separation of PBMCs by standard Ficoll-Paque density gradient centrifugation, according to previously described technique [32,33]. Total cellular nucleic acid was extracted from PBMCs using the TRI Reagent (Molecular Research Center, Clicinnati, OH) according to the manufacturer's protocol, and stored at -70°C for RT-PCR analysis.

Real-time RT-PCR
Total RNA (1 μg) was reverse transcribed using the RT system (Promega, Madison, WI) for 1 h at 42°C. 1.5 μL cDNA was subjected as a template for real-time PCR quantification. Using the power SYBR ® Green PCR Master Mix (Life Technologies, California), messenger RNAs (mRNAs) in PBMCs were detected. The level of glyceraldehydes-3-phosphate dehydrogenase (GAPDH) mRNA was used as an endogenous reference to normalize the quantities of target mRNA. The primers used for the detection of the mRNAs ( Table 2) were designed as previously described [12,22]. The PCR results were analyzed by the 2 -ΔΔCT methods [34].

Statistical analysis
Where appropriate, data were expressed as mean ± standard deviations (SD). For comparison of the means of two groups, statistical significance was assessed by Student's t paired (prior to versus after cART) or unpaired (HIV-1-infected subjects versus HIV-1-uninfected subjects) test. Statistical analyses were performed with GraphPad Instat statistical software (San Diego, CA). Statistical significance was defined as p < 0.05.