Enhanced Expression of Two ISGs as Potential Biomarkers of Prognosis in HIV-Infected Immunological Non-Responders

Abstract

Objective:

No novel biomarkers are currently available for evaluating immune reconstitution among people living with HIV/AIDS (PLWH) receiving combined antiretroviral treatment (cART). The objective of this study was to analyze the expression patterns of interferon-stimulating genes (ISGs) in PLWH with the aim of identifying potential biomarkers for immune reconstitution.

Methods:

Study samples were collected from 105 PLWH, including 47 immunological non-responders (INRs) and 58 immunological responders (IRs). The expression of eight ISGs in the peripheral blood in INRs and IRs were detected by RT-qPCR. Expression differences between groups were analyzed with the Mann-Whitney U test, and a logistic regression model was developed to predict immune reconstitution.

Results:

Among eight ISGs, the expression levels of IFI27 and IFI6 were significantly higher in INRs than IRs (P=0.001 and 0.005, respectively). The model combining age, CD4⁺ T/CD8⁺ T ratio, IFI27, and IFI6 had the highest diagnostic value (AUC=0.836), with an optimal cut-off value of 0.6, sensitivity of 60.5%, and specificity of 98.1%. No significant change in the expression of IFI27 and IFI6 was observed in samples collected 3 years apart (P=0.1232 and 0.4877, respectively), and the model score negatively correlated with ΔCD4⁺ T cells (r=−0.2888, P=0.0465).

Conclusions:

Enhanced expression of IFI27 and IFI6 in INRs are important characteristics that may serve as biomarkers of immune reconstitution after cART. The combination of age, CD4⁺ T/CD8⁺ T ratio, IFI27, and IFI6 was highly effective in discriminating INRs.

Main article text

INTRODUCTION

HIV invades primarily CD4⁺ T lymphocytes and subsequently leads to progressive immunodeficiency, opportunistic infections, and cancers [1]. cART effectively suppresses HIV replication, and decreases cellular and soluble activation markers, thus resulting in immune recovery in a high proportion of individuals [2]. However, with increasing accessibility and use of antiretroviral treatment, some HIV-infected individuals, referred to as INRs, achieve viral suppression but not optimal CD4⁺ T cell recovery. To date, no formal standard definition of INRs exists. Given the heterogeneity among study populations and discrepancies in definitions, the reported prevalence of INRs varies from 10% to 40% [3]. A previous study has indicated that the relative risk of clinical progression (AIDS-defined event or death) of patients having a CD4⁺ T cell increase below 100 cells/μL per year is 13.3-fold higher than that of patients above this threshold [4]. Currently, the lack of appropriate clinical indicators to evaluate immune recovery status among infected patients in early stages after antiretroviral therapy has become an important factor restricting progress in this field.

Recent studies have highlighted the deleterious effects of IFN in chronic infections [5]. During chronic HIV infection, high IFN-I signaling is correlated with immune activation [6], poor treatment-associated immune reconstitution [7–9], and disease progression [5,10,11]. In a humanized mouse model infected with HIV, blocking of type I interferon with antiretroviral therapy inhibits non-specific immune activation and restores immune cell function [12]. Given that these results suggest that improved understanding of the IFN response in chronic HIV infection may yield new insights into the immune response under ART, expression of IFN-1 pathway molecules might potentially be used to evaluate and predict immune reconstitution in HIV infections during early stages of antiretroviral therapy.

However, in persistent infections , the burst of IFN-I production is transient, and levels subsequently decrease to those observed in naïve hosts, although interferon-dependent ISG signatures remain [13]. Chronic ISG signaling is also a consistent signature of blunted homeostasis in CD4⁺ T cells during ART [5]. Although the mechanism maintaining interferon-dependent ISG signaling is not precise, ISG molecules can be used as biomarkers of interferon pathway activation. In our previous work in a humanized mouse model, we identified eight potential immune-reconstructing ISGs through flow cytometry and RNA-seq analysis: CXCL10, IFI44, IFIT1, IFIT2, MX1, MX2, IFI27, and IFI6. We hypothesized that these ISGs might aid in understanding the mechanisms of poor CD4⁺ T cell recovery and immune reconstruction in HIV infection.

To properly characterize these ISGs in INRs and IRs, we first established an RT-qPCR method for quantifying ISGs in peripheral blood. We performed quantitative ISG assays in INRs and IRs receiving ART with these approaches. Our study is the first to use qPCR to explore the association between ISG expression and CD4⁺ T cell recovery. These analyses were designed to evaluate the application value of ISGs as biomarkers to assess immune reconstitution and to inform subsequent research.

MATERIALS AND METHODS

Participants

Participants were recruited from Yunnan Dehong prefecture, where CD4 count detection is required before treatment and once per year after antiviral treatment. The criteria for participant selection were as follows: patients with pre-cART CD4⁺ T cell count ≤200 cells/μL, who had received continuous and regular cART for more than 2 years and showed complete viral suppression (VL <200 copies/mL). On the basis of the above inclusion criteria, 105 HIV-infected patients were selected from the treatment pool from the year 2015, 52 of whom also had follow-up samples collected in 2018. A total of 157 lymphocyte-enriched samples were eventually included in this study. We collected patients’ demographic information, including age, sex, CD4⁺ T cell count, CD8⁺ T cell count, and ART time.

IRs were defined as patients with CD4⁺ T cell counts >500 cells/μL after receiving ART, whereas patients with a CD4⁺ T cell count consistently below 200 cells/μL were classified as INRs. The study and all research protocols were approved by the Institutional Review Board of the National Center for AIDS/STD Prevention and Control, Chinese Center for Disease Control and Prevention (X191030588).

Total mRNA extraction

The lymphocyte enrichment samples from patients were isolated from whole blood with density gradient centrifugation and stored at −80°C. Total RNA was purified from lymphocyte enrichment samples with an RNA Simple Total RNA Kit (TIANGEN, DP419, China) based on phenol and guanidine thiocyanate. PBMC samples stored at −80°C were removed, and a 3× volume of Buffer RZ was immediately added. The lysates were then completely disrupted and homogenized at room temperature. RZ Lysis Reagent, designed to facilitate tissues lysis and inhibit RNases, was used to prevent RNA degradation, and remove genomic DNA and protein from cells. The rest of the procedure was performed in strict accordance with the manufacturer’s instructions.

Primer and probe design

The complete transcript variants’ mRNA sequences for each ISG were downloaded from the NCBI database and aligned with SnapGene v4.2.1 to identify conserved regions. GAPDH was used as the internal control gene. All primers and probes for each ISG and the internal reference gene were designed in Oligo 7. The specificity of each primer and probe was verified through BLAST searching against the NCBI database. The primers and probes designed in this study were synthesized by Sangon Biotech Co., Ltd (Beijing, China) and are summarized in S1 Table.

RT-qPCR

We established a dual RT-qPCR assay to quantify the gene expression of the eight ISGs and the internal control gene GAPDH in one tube. The RT-qPCR reactions used a QuantiNova Probe RT-PCR Kit (Qiagen 208354) and were performed on a Bio-Rad CFX96 system as follows: after 10 min of reverse transcription at 45°C and 5 min of activation of DNA polymerase at 95°C, amplification was performed for 40 cycles consisting of a denaturation step at 95°C for 5 s and combined annealing/extension step at 60°C for 30 s. S1 Fig shows the standard curve and amplification plot for the ISGs and GAPDH. S1 Table lists the primers used for detection.

ISG expression data were analyzed with the 2^−ΔCt method. The Ct value of GAPDH was used for normalization of transcript abundance. The relative expression values represent the fold change with respect to GAPDH expression [14].

Synthesis of RNA transcripts and generation of standard curves

RNA standards were amplified from the conserved regions of ISGs with primers containing an upstream T7 promoter sequence and were transcribed in vitro with the T7 RiboMAX™ Express Large Scale RNA Production System (Promega P1320). The synthetic RNA transcripts were then purified and quantified on a NanoDrop Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). To determine whether the amplification efficiencies of ISGs and GAPDH were similar, the in vitro-derived RNA transcripts were 10-fold diluted in the range of 10⁸ to 10⁴ RNA copies/μL, and calibration curves were generated.

Statistical analysis

Median and interquartile range (IQR) were used to describe variables with non-normal distribution. Nonparametric tests (Mann-Whitney U test for continuous variable and chi-square (X ²) or Fisher’s exact test for categorical variables) were used to compare variables of INRs and IRs. The comparison of variables was performed with a Wilcoxon matched-pairs signed-rank test for paired data. Correlation analysis used Spearman correlation and was visualized with Bitmap. All statistically significant variables were candidates for further logistic regression analyses to construct a multivariate logistic regression model. Receiver operating characteristic (ROC) curve analysis was used to estimate the ability to identify the immune reconstruction in HIV-infected individuals. All tests were two-tailed, and a P-value < 0.05 was considered statistically significant. All statistical analyses were performed in SPSS 22.0, GraphPad Prism 8, and Origin2021.

RESULTS

Demographic characteristics of study participants

A total of 47 INRs (31 men) and 58 IRs (31 men) were included (Table 1). The INRs were significantly younger in age than the IRs (P=0.004). The median baseline CD4⁺ T/CD8⁺ T ratio of INRs was less than that of IRs (P=0.034), at 0.11 [0.05–0.19] and 0.17 [0.07–0.30], respectively. However, no significant differences were observed in other variables (sex, baseline CD4⁺ T cell count, baseline CD8⁺ T cell count, and ART period) between groups (Table 1).

TABLE 1 |

Characteristics of the patient subgroup.

Variable	Patient subgroup		P-value
	INR (n=47)	IR (n=58)
Sex (male:female)	31:16	31:27	0.088
Age	46 (37–52)	39 (35–46)	0.004
Baseline CD4+ T (cells/μL)	82 (27–152)	135 (54–170)	0.084
Baseline CD8+ T (cells/μL)	563 (401–1188)	560 (345–750)	0.260
Baseline CD4+ T/CD8+ T ratio	0.11 (0.05–0.19)	0.17 (0.07–0.30)	0.034
ART time	5.50 (3.76–7.22)	5.97 (4.26–7.46)	0.657

Standard curve

To evaluate the amplification efficiencies of the quantitative real-time RT-PCR assay established for the eight ISGs and GAPDH, we used ten-fold serial dilutions of synthetic RNA transcripts to generate calibration curves. All assays showed favorable amplification efficiencies exceeding 90%, and a strong linear correlation (r²>0.995) between the Ct values and each ISG transcript within at least a 5-log range from 10⁴ to 10⁸ copies/μL (S1 Fig).

Correlation analysis

To enhance understanding of the ISGs, we conducted a correlation analysis. Fig 1 illustrates a visual representation of the correlation analysis among the ISGs, with red and blue colors indicating positive and negative correlations, respectively. Additionally, the intensity of the color corresponds to the magnitude of the correlation coefficient. On the basis of Spearman’s correlation coefficient (r), values ranging from 0 to 0.30, 0.30 to 0.50, 0.50 to 0.70, and 0.70 to 1.00 were indicative of “poor,” “fair” or “moderate,” “good,” and “strong” correlations, respectively. As depicted in Fig 1, the ISGs were essentially positively correlated, and a notable positive correlation was observed among IFI44, IFIT1, IFIT2, MX1, MX2, and IFI6.

FIGURE 1 |

Bitmap of the correlation analysis among the eight ISGs in 105 samples. Red and blue denote positive and negative correlations. *denotes statistically correlation. P<0.05 was considered statistically significant.

Differential expression of ISGs in INRs and IRs

We quantified the expression of ISGs in PBMC samples. Unexpectedly, the levels of IFI27 and IFI6 were higher in INRs than IRs (P=0.001 and 0.005, respectively) (Fig 2), but no significant differences were observed between INRs and IRs in other ISGs.

FIGURE 2 |

Relative expression of eight ISGs in two patient subgroups.

Additionally, to investigate the correlation of IFI27 and IFI6 expression with immune recovery, we performed correlation analysis, which indicated that IFI27 and IFI6 did not correlate with baseline CD4⁺ T cell count. In contrast, the expression levels of IFI27 and IFI6 both negatively correlated with CD4⁺ T cell count, and showed statistically significant differences (r=−0.267, P=0.006; r=−0.191, P=0.051, respectively) (Fig 3).

FIGURE 3 |

A. Correlation analysis between baseline CD4⁺ T cell counts and IFI27 expression (Spearman correlation test). r=−0.061 P=0.540. B. Correlation analysis between baseline CD4⁺ T cell counts and IFI27 expression (Spearman correlation test). r=−0.062 P=0.532. C. Correlation analysis between CD4⁺ T cell counts and IFI27 expression (Spearman correlation test). r=−0.267 P=0.006. D. Correlation analysis between CD4⁺ T cell counts and IFI6 expression (Spearman correlation test). r=−0.191 P=0.051. The brown dot represents INR, and the blue dot represents IR.

We further performed expression stratification analysis, and divided patients into high, moderate, or low expression groups according to IFI27 and IFI6 expression level. In comparison to INR, a greater proportion of IR was observed in groups with low expression of IFI6 and IFI27 (Tables 2A and 2B). The proportion of participants with a relative expression of IFI27 exceeding 3.0 was greater among INRs (12.8%) than IRs (0.0%). Similarly, the rate of relative expression of IFI6 above 0.8 in INRs was 17.0%, a percentage higher than that among IRs (0.0%).

TABLE 2A |

Relative IFI27 expression distribution in INRs and IRs.

Relative expression of IFI27	INRs (%)	IRs (%)	X 2	P-value
Low expression (<1.0)	34 (72.3)	57 (98.3)	15.165	0.001
Moderate expression (1.0–3.0)	7 (14.9)	1 (1.7)
High expression (>3.0)	6 (12.8)	0 (0.0)
Total	47 (100.0)	58 (100.0)

TABLE 2B |

Relative IFI6 expression distribution in INRs and IRs.

Relative expression of IFI6	INRs (%)	IRs (%)	X 2	P-value
Low expression (<0.4)	34 (72.3)	54 (93.1)	12.127	0.001
Moderate expression (0.4–0.8)	5 (10.6)	4 (6.9)
High expression (>0.8)	8 (17.0)	0 (0.0)
Total	47 (100.0)	58 (100.0)

Logistic regression analysis and evaluation of the model

Table 1 and Fig 2 show that compared to INRs, IRs were younger, had lower CD4⁺ T/CD8⁺ T ratios, and overexpressed IFI6 and IFI27. To estimate the ability of different variables to distinguish between INRs and IRs, we conducted ROC analysis of ratio, age, IFI27 expression, and IFI6 expression. The area under the ROC curve (AUC) values were 0.6782, 0.6645, 0.6807, and 0.6544, respectively.

Additionally, a multivariate logistic regression model incorporating the above four variables was developed (Table 3). The regression equation was as follows: logit(p)=−4.294–5.989X₁+0.090X₂+1.566X₃+2.911X₄. The Hosmer-Lemeshow fit index was 6.207 (P>0.05), thus indicating that the equation was well matched. We calculated the scores of the regression equation for each individual and performed ROC analysis. Combining the four variables resulted in higher AUCs than those for each variable alone across all models. The AUC was 0.836 (95% CI: 0.752–0.921) (P<0.0001), and the optimal cut-off value was 0.600, whereas the model showed 98.1% specificity, 60.5% sensitivity, and a maximum Youden index of 0.586 (Fig 4).

FIGURE 4 |

ROC curves for multi-factor regression analysis of four significant variables and the combined model.

TABLE 3 |

Multivariate logistic regression analysis of significant variables.

Variable	b	SE	Ward	P	OR	LL	UL
Constant	−4.294	1.400	3.067	0.002	0.014
Ratio	−5.989	2.519	2.377	0.017	0.003	0.000	0.196
Age	0.090	0.029	3.081	0.002	1.094	1.038	1.165
IFI27	1.566	0.785	1.994	0.046	4.787	1.298	28.440
IFI6	2.911	1.537	1.895	0.058	18.380	1.995	746.100

Ratio=CD4⁺ T/CD8⁺ T ratio; LL=lower limit; OR=odds ratio; SE=standard error; UL=upper limit.

Longitudinal analysis

We collected paired samples from 52 individuals, including 20 INRs and 32 IRs, in 2015 and 2018, respectively. A paired test indicated no statistical difference in IFI27 and IFI6 expression between the timepoints (P=0.1232, 0.4877, respectively), thus implying that the expression of these two ISGs was relatively stable (Fig 5).

FIGURE 5 |

Effect of ART on ISGs. A. Longitudinal analysis of the change in IFI27 in PLWH after 3 years on ART (P=0.1232). B. Longitudinal analysis of the change in IFI6 in PLWH after 3 years on ART (P=0.4877). The brown dot represents INR, the blue dot represents IR, and the black dot represents individuals with a CD4⁺ T cell count of 201–350 cells/μL.

Moreover, the scores calculated with the combination model negatively correlated with ΔCD4⁺ T cells at the 3-year interval (r=−0.2888, P=0.0465). Among INRs with scores ≤0.6, 71.42% (5/7) of individuals achieved an increase in CD4⁺ T cell count after 3 years (65–1220). Among the 20 INRs, all scores were ≤0.600, except for three cases in which the score could not be calculated, because of a missing CD4⁺ T/CD8⁺ T ratio; 72.4% (21/29) of the infected patients achieved an increase in CD4⁺ T cell count 3 years later.

DISCUSSION

Our data showed that IRs were younger and had a higher baseline CD4⁺ T/CD8⁺ T ratio compared to INRs, in agreement with findings from prior studies [15–17]. Previous studies have also indicated that multiple factors are associated with poor recovery of CD4⁺ T cells during HAART, including co-infection [18–21], HAART regimen [22], decreased CD4⁺ T cell proliferation [15], and host genetic variation [23]. However, none of these independent factors can fully explain the mechanisms underlying immunological non-response. The molecular mechanism underlying poor immune reconstruction remains to be further explored.

We developed RT-qPCR assays for immune reconstruction associated with the eight ISGs, then performed ISG expression assays on PBMCs of INRs and IRs. The expression and correlations for the eight ISGs were analyzed. A positive correlation was observed among the expression of most ISGs, thus verifying that these ISGs are coordinately expressed and share common regulation by the IFN signaling pathway. Notably, we observed higher expression of two ISGs in INRs than IRs. These two ISGs, IFI6 and IFI27, belong to the FAM14 family [24], and are mitochondrial proteins with distinct functions in apoptosis. The FAM14 family is widely expressed in various cells/tissues with multiple biological functions. Overexpression of either IFI6 or IFI27 inhibits the growth and migration of tongue squamous cell carcinoma cells [25].

Notably, growing evidence indicates that IFI27 might play a role in restoring immune reconstruction. IFI27 expression is relatively high in the CD4⁺ T cells in PLWH with low CD4⁺ T cell counts [8], whereas IFI27 expression is higher in the monocytes in HIV infected individuals with a high HIV-1 viral load [26]. Liu et al., through a combined DEG analysis, WGCNA hub gene analysis, and meta-DEG approach, have reported that IFI27 is consistently upregulated in INRs across multiple datasets and conditions, particularly in PBMC samples [27]. These findings were supported by our data. Of note, IFI6 expression was also upregulated among INRs, whereas this ISG was not previously described in studies associated with immune reconstruction. This study provides the first characterization of IFI6 expression in PLWH. A previous study has confirmed that enhanced IFI6 expression correlates with the phenotype of invasive disease and poor prognosis in patients with esophageal squamous cell carcinoma [28], whereas, to date, the role and the driving factor of IFI6 expression in infection and immune reconstruction remained to be explored.

In addition, we further examined the relationship between ISGs and CD4⁺ T cell recovery, and found that the expression of both ISGs was significantly negatively correlated with the CD4⁺ T cell counts of PLWH. Through analysis of expression distribution, we determined that IFI6 (relative expression level >0.8) and IFI27 (relative expression level >3.0) had significant efficacy in predicting the risk of poor immune recovery. As illustrated in Fig 5, we also observed the clear opposite change trends in the partial IRs, characterized by decreased IFI27 and IFI6 expression level but increased CD4⁺ T count. These results underscore the crucial association between enhanced ISGs expression and poor immune reconstitution, suggesting that these two ISGs might hold significant diagnostic value for poor immune recovery and could serve as a potential marker for distinguishing INRs in PLWH. However, the specific mechanisms underlying the interactions among IFN, ISG, and CD4⁺ T cell recovery remain to be further explored.

The AUC values for the statistically significant variables (CD4⁺ T/CD8⁺ T ratio, age, IFI27, and IFI6) were all below 0.7, thus suggesting limited discriminatory ability to distinguish INRs from IRs. Previous models/methods based on factors including the baseline CD4⁺ T cell count and baseline CD4⁺ T/CD8⁺ T ratio, have been reported to be unable to predict immune reconstruction [29]. However, the combination of these variables showed the most promise in discriminating INRs from HIV-infected individuals. ROC curves indicated the diagnostic ability of the combined model, which might serve as an indicator of the immune reconstruction after treatment. The model’s sensitivity and specificity were superior to those of baseline CD4⁺ T/CD8⁺ T ratio, age, IFI6, and IFI27. Longitudinal analysis indicated no significant differences in IFI6 and IFI27 over a 3-year period, thereby suggesting that expression of these ISGs is fairly stable in vivo; therefore, these ISGs could serve as indicators of long-term status. More importantly, a negative correlation was observed between the score calculated with the combined model and ΔCD4⁺ T cells, thus indicating that the combined model, based on stably expressed ISGs, could potentially early predict immune reconstitution of PLWH. These findings from the present study provide a platform that may potentially be used to further validate the efficacy of our models in larger sample sizes. Prediction tools for immune reconstruction play a critical role in facilitating early clinical intervention. Although no precise mechanism is understood to explain poor immune reconstitution, Tripterygium wilfordii Hook F, IL-7, and Wenshen Jianpi recipe have shown efficacy in facilitating CD4⁺ T cell recovery [30–32], thereby aiding in immune reconstitution. Achieving complete immune reconstruction significantly decreases the morbidity and mortality associated with both AIDS and non-AIDS-related events, such as metabolic syndrome, liver diseases, nephropathy, cardiovascular disease, non-AIDS-related malignancies, and HIV-1-related neurocognitive disorder [3].

This study has several limitations. First, the number of participants in this study was relatively small, and no ISG expression data before the initiation of treatment was available. Second, the use of whole PBMCs, rather than specific cell subtypes, can substantially simplify the procedure and aid in implementation of ISGs assays, whereas the expression of ISGs may vary among cell subgroups. Third, although we identified enhanced expression of two ISGs in poor CD4⁺ T cell recovery, the hierarchical processes and mechanism between them remains to be fully elucidated.

In summary, we developed simple and easy-to-implement ISGs assays and built a high-efficiency combined module to discriminate INRs among PLWH by using two ISGs. Although the molecular mechanisms underlying the interactions among these two ISG and poor CD4⁺ T cell recovery remains to be further explored, the expression of these two ISG in INRs and IRs may support future research, and contribute to understanding of the detailed functions of IFI6 and IFI27 in the prognosis of immune reconstruction.

CONFLICTS OF INTEREST

The members of the writing committee declare that they have no conflicts of interest.

SUPPLEMENTARY MATERIAL

Supplementary Material can be downloaded from https://zoonoses-journal.org/wp-content/uploads/2024/08/zoonoses20240016_Suppl.pdf

REFERENCES

1. Lucas S, Nelson AM. HIV and the spectrum of human disease. J Pathol. 2015. Vol. 235(2):229–241

2. Battegay M, Nüesch R, Hirschel B, Kaufmann GR. Immunological recovery and antiretroviral therapy in HIV-1 infection. Lancet Infect Dis. 2006. Vol. 6(5):280–287

3. Yang X, Su B, Zhang X, Liu Y, Wu H, Zhang T. Incomplete immune reconstitution in HIV/AIDS patients on antiretroviral therapy: challenges of immunological non-responders. J Leukoc Biol. 2020. Vol. 107(4):597–612

4. Piketty C, Weiss L, Thomas F, Mohamed AS, Belec L, Kazatchkine MD. Long-term clinical outcome of human immunodeficiency virus-infected patients with discordant immunologic and virologic responses to a protease inhibitor-containing regimen. J Infect Dis. 2001. Vol. 183(9):1328–1335

5. Deeks SG, Odorizzi PM, Sekaly RP. The interferon paradox: can inhibiting an antiviral mechanism advance an HIV cure? J Clin Invest. 2017. Vol. 127(1):103–105

6. Hardy GA, Sieg S, Rodriguez B, Anthony D, Asaad R, Jiang W, et al.. Interferon-α is the primary plasma type-I IFN in HIV-1 infection and correlates with immune activation and disease markers. PLoS One. 2013. Vol. 8(2):e56527

7. Le Saout C, Hasley RB, Imamichi H, Tcheung L, Hu Z, Luckey MA, et al.. Chronic exposure to type-I IFN under lymphopenic conditions alters CD4 T cell homeostasis. PLoS Pathog. 2014. Vol. 10(3):e1003976

8. Fernandez S, Tanaskovic S, Helbig K, Rajasuriar R, Kramski M, Murray JM, et al.. CD4+ T-cell deficiency in HIV patients responding to antiretroviral therapy is associated with increased expression of interferon-stimulated genes in CD4+ T cells. J Infect Dis. 2011. Vol. 204(12):1927–1935

9. Dunham RM, Vujkovic-Cvijin I, Yukl SA, Broadhurst MJ, Loke P, Albright RG, et al.. Discordance between peripheral and colonic markers of inflammation during suppressive ART. J Acquir Immune Defic Syndr. 2014. Vol. 65(2):133–141

10. Rotger M, Dalmau J, Rauch A, McLaren P, Bosinger SE, Martinez R, et al.. Comparative transcriptomics of extreme phenotypes of human HIV-1 infection and SIV infection in sooty mangabey and rhesus macaque. J Clin Invest. 2011. Vol. 121(6):2391–2400

11. Tenorio AR, Zheng Y, Bosch RJ, Krishnan S, Rodriguez B, Hunt PW, et al.. Soluble markers of inflammation and coagulation but not T-cell activation predict non-AIDS-defining morbid events during suppressive antiretroviral treatment. J Infect Dis. 2014. Vol. 210(8):1248–1259

12. Cheng L, Ma J, Li J, Li D, Li G, Li F, et al.. Blocking type I interferon signaling enhances T cell recovery and reduces HIV-1 reservoirs. J Clin Invest. 2017. Vol. 127(1):269–279

13. Wilson EB, Yamada DH, Elsaesser H, Herskovitz J, Deng J, Cheng G, et al.. Blockade of chronic type I interferon signaling to control persistent LCMV infection. Science. 2013. Vol. 340(6129):202–207

14. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods. 2001. Vol. 25(4):402–408

15. Maartens G, Celum C, Lewin S. HIV infection: epidemiology, pathogenesis, treatment, and prevention. Lancet. 2014. Vol. 384(9939):258–271

16. Kroeze S, Ondoa P, Kityo CM, Siwale M, Akanmu S, Wellington M, et al.. Suboptimal immune recovery during antiretroviral therapy with sustained HIV suppression in sub-Saharan Africa. AIDS. 2018. Vol. 32(8):1043–1051

17. Ahn MY, Jiamsakul A, Khusuwan S, Khol V, Pham TT, Chaiwarith R, et al.. The influence of age-associated comorbidities on responses to combination antiretroviral therapy in older people living with HIV. J Int AIDS Soc. 2019. Vol. 22(2):e25228

18. Anderson M, Gaseitsiwe S, Moyo S, Thami KP, Mohammed T, Setlhare D, et al.. Slow CD4(+) T-cell recovery in human immunodeficiency virus/hepatitis B virus-coinfected patients initiating truvada-based combination antiretroviral therapy in Botswana. Open Forum Infect Dis. 2016. Vol. 3(3):ofw140

19. Potter M, Odueyungbo A, Yang H, Saeed S, Klein MB. Impact of hepatitis C viral replication on CD4+ T-lymphocyte progression in HIV-HCV coinfection before and after antiretroviral therapy. AIDS. 2010. Vol. 24(12):1857–1865

20. Gómez-Mora E, Massanella M, García E, Giles D, Bernadó M, Urrea V, et al.. Elevated humoral response to cytomegalovirus in HIV-infected individuals with poor CD4+ T-cell immune recovery. PLoS One. 2017. Vol. 12(9):e0184433

21. Chen M, Wong WW, Law MG, Kiertiburanakul S, Yunihastuti E, Merati TP, et al.. Hepatitis B and C Co-infection in HIV patients from the TREAT Asia HIV observational database: analysis of risk factors and survival. PLoS One. 2016. Vol. 11(3):e0150512

22. Edwards JK, Cole SR, Hall HI, Mathews WC, Moore RD, Mugavero MJ, et al.. Virologic suppression and CD4+ cell count recovery after initiation of raltegravir or efavirenz-containing HIV treatment regimens. AIDS. 2018. Vol. 32(2):261–266

23. García M, Jiménez-Sousa MA, Blanco J, Restrepo C, Pacheco YM, Brochado-Kith Ó, et al.. CD4 recovery is associated with genetic variation in IFNγ and IL19 genes. Antiviral Res. 2019. Vol. 170:104577

24. Parker N, Porter AC. Identification of a novel gene family that includes the interferon-inducible human genes 6-16 and ISG12. BMC Genomics. 2004. Vol. 5(1):8

25. Xu L, Zu T, Li T, Li M, Mi J, Bai F, et al.. ATF3 downmodulates its new targets IFI6 and IFI27 to suppress the growth and migration of tongue squamous cell carcinoma cells. PLoS Genet. 2021. Vol. 17(2):e1009283

26. Patro SC, Pal S, Bi Y, Lynn K, Mounzer KC, Kostman JR, et al.. Shift in monocyte apoptosis with increasing viral load and change in apoptosis-related ISG/Bcl2 family gene expression in chronically HIV-1-infected subjects. J Virol. 2015. Vol. 89(1):799–810

27. Liu X, Lin L, Lu L, Li X, Han Y, Qiu Z, et al.. Comparative transcriptional analysis identified characteristic genes and patterns in HIV-infected immunological non-responders. Front Immunol. 2022. Vol. 13:807890

28. Liu Z, Gu S, Lu T, Wu K, Li L, Dong C, et al.. IFI6 depletion inhibits esophageal squamous cell carcinoma progression through reactive oxygen species accumulation via mitochondrial dysfunction and endoplasmic reticulum stress. J Exp Clin Cancer Res. 2020. Vol. 39(1):144

29. Raffi F, Le Moing V, Assuied A, Habak S, Spire B, Cazanave C, et al.. Failure to achieve immunological recovery in HIV-infected patients with clinical and virological success after 10 years of combined ART: role of treatment course. J Antimicrob Chemother. 2017. Vol. 72(1):240–245

30. Li T, Xie J, Li Y, Routy JP, Li Y, Han Y, Qiu Z, et al.. Tripterygium wilfordii Hook F extract in cART-treated HIV patients with poor immune response: a pilot study to assess its immunomodulatory effects and safety. HIV Clin Trials. 2015. Vol. 16(2):49–56

31. Thiebaut R, Jarne A, Routy JP, Sereti I, Fischl M, Ive P, et al.. Repeated cycles of recombinant human interleukin 7 in HIV-Infected patients with low CD4 T-cell reconstitution on antiretroviral therapy: results of 2 phase II multicenter studies. Clin Infect Dis. 2016. Vol. 62(9):1178–1185

32. Zhuang T, Jian W, Xin C, Yonghong LI, Yuguang Y, Ao Z, et al.. Wenshen Jianpi recipe induced immune reconstruction and redistribution of natural killer cell subsets in immunological non-responders of human immunodeficy virus/acquired immune deficiency syndrome: a randomized controlled trial. J Tradit Chin Med. 2022. Vol. 42(5):795–802