Interferon-stimulated genes (ISGs) constitute the core effector layer for establishing the cellular antiviral state. After viral RNA is recognized by pattern recognition receptors such as RIG-I and MDA5, type I/type III interferon production can be induced through MAVS, TBK1, IRF3/IRF7, and NF-κB. These interferons then drive ISG expression through the JAK-STAT pathway. ISG-related regulatory factors such as IFI35 can participate in fine regulation of the intensity, duration, and inflammatory output of the interferon response.

Keywords: interferon-stimulated genes; RIG-I; MDA5; IFI35; ISG; antiviral immunity; type I interferon; JAK-STAT

1 Research Logic of Interferon-Stimulated Genes and Antiviral Immunity

1.1 Functional positioning of ISGs

(1) Execution of antiviral effects

ISGs are a group of effector genes induced after interferon signaling is activated. Their functions cover multiple stages, including restriction of viral entry, viral RNA degradation, inhibition of protein translation, interference with viral replication complexes, suppression of particle assembly, and immune marking of infected cells. Common ISGs include ISG15, MX1, OAS1, OAS2, IFIT1, IFIT2, IFIT3, RSAD2, PKR/EIF2AK2, IFITM1/2/3, and BST2.

(2) Signal amplification and feedback regulation

ISGs are not only terminal antiviral effector molecules; they can also participate in upstream signal regulation. Some ISGs enhance viral nucleic acid recognition or promote interferon responses, whereas others limit excessive inflammation and persistent JAK-STAT activation. In antiviral immunity research, both the effector functions and feedback-regulatory functions of ISGs should be considered.

(3) Cell-type dependence

ISG expression profiles differ substantially among cell types. Epithelial cells, macrophages, dendritic cells, hepatocytes, neurons, and tumor cells respond differently to viral infection, dsRNA mimics, type I interferons, or type III interferons. Experimental design should not rely on a single ISG. Instead, marker combinations should be selected according to cell type, virus type, and infection stage.

1.2 Basic axis of antiviral signaling

(1) Viral RNA recognition

RIG-I and MDA5 are key RLR (RIG-I-like receptor) members in cytosolic RNA virus recognition. RIG-I preferentially recognizes viral RNA with 5′-triphosphate ends, short double-stranded regions, or specific terminal structures. MDA5 is more commonly associated with long double-stranded RNA or replication intermediates. Although they recognize different RNA substrates, both can connect downstream antiviral signaling through MAVS.

(2) Interferon induction

After RIG-I or MDA5 activation, MAVS forms a signaling platform on the outer mitochondrial membrane and further activates TBK1, IKKε, IRF3, IRF7, and NF-κB. IRF3/IRF7 promotes transcription of IFNB1, IFNA subtypes, and some interferon-related genes, whereas NF-κB drives expression of inflammatory cytokines and chemokines such as TNF, IL6, CXCL10, and CCL5.

(3) ISG expression execution

Secreted IFN-β, IFN-α, or IFN-λ binds to the corresponding receptors and activates JAK1, TYK2, STAT1, and STAT2. STAT1, STAT2, and IRF9 form the ISGF3 complex, which enters the nucleus and induces ISRE-dependent ISG expression. At this stage, cells enter an antiviral alert state and restrict viral replication and spread.

Table 1 RIG-I/MDA5-IFN-ISG Antiviral Signaling Framework

2 Division of Labor Between RIG-I and MDA5 in Viral RNA Recognition

2.1 RIG-I signaling features

(1) Recognition of short RNA and 5′-triphosphate structures

RIG-I is encoded by DDX58 and functions as a cytosolic RNA sensor. It commonly recognizes RNA with 5′-triphosphate or 5′-diphosphate ends, short double-stranded structures, or specific viral RNA conformations. Many negative-strand RNA viruses, replication intermediates of some positive-strand RNA viruses, and synthetic 5′ppp RNA can induce RIG-I activation.

(2) CARD domain-mediated MAVS activation

After binding viral RNA, RIG-I undergoes a conformational change that exposes its N-terminal CARD domains, allowing interaction with MAVS. Activated MAVS further recruits downstream kinase complexes and initiates IRF3/IRF7 and NF-κB pathway activation.

(3) Key points for experimental interpretation

Increased RIG-I expression does not equal RIG-I activation. Studies should evaluate RIG-I protein conformation, downstream MAVS signaling, p-TBK1, p-IRF3, IFNB1, and ISG expression together. If RIG-I increases but IFNB1 and ISGs show no obvious change, this may indicate a pre-activated state, feedback induction, or insufficient recognition substrate.

2.2 MDA5 signaling features

(1) Recognition of long dsRNA or viral replication intermediates

MDA5 is encoded by IFIH1 and is commonly associated with recognition of long double-stranded RNA, viral replication intermediates, and some positive-strand RNA viruses. High-molecular-weight Poly(I:C) is often used to mimic MDA5-related dsRNA stimulation, but the actual response is still influenced by delivery method, cell type, and endosomal TLR3 background.

(2) Oligomerization and signaling threshold

MDA5 recognition of long dsRNA usually involves assembly and oligomerization along the RNA molecule. Its signaling initiation has a threshold-like feature. Low-level or short-fragment dsRNA may not be sufficient to induce strong MDA5 signaling. High-dose stimulation may trigger strong interferon responses and cytotoxicity, so cell viability should be assessed in parallel.

(3) Complementary relationship with RIG-I

RIG-I and MDA5 differ in their preferences for virus types and RNA conformations, but cross-regulation or stage-specific cooperation can occur in many viral infections. During early infection, RIG-I may respond more rapidly to specific viral RNA structures. During middle or late replication stages, MDA5 may contribute more strongly to accumulated dsRNA intermediates. Knockdown, knockout, or specific stimulant comparison should be used to determine their respective roles.

2.3 Interpretation of RIG-I/MDA5 results

(1) Interpretation by virus type

Influenza virus, VSV, and SeV are commonly used in RIG-I-related signaling studies. EMCV, some small RNA viruses, and long dsRNA mimics are commonly used in MDA5-related studies. Different viruses differ in replication mode, RNA structure, and immune evasion proteins, so one viral model cannot represent all RLR signaling.

(2) Interpretation by stimulation method

Transfection of 5′ppp RNA, short dsRNA, or extracted viral RNA is more suitable for evaluating RIG-I. Cytosolic delivery of high-molecular-weight Poly(I:C) is more suitable for evaluating MDA5. If Poly(I:C) is directly added to the culture medium without transfection, the result may involve more TLR3 or endocytic pathways.

(3) Interpretation by downstream signaling

Both RIG-I and MDA5 signal through MAVS into the TBK1-IRF3 and NF-κB pathways. Therefore, increased IFNB1 or ISG expression alone cannot distinguish the specific upstream receptor. Interpretation should integrate DDX58 knockdown/knockout, IFIH1 knockdown/knockout, MAVS deficiency, stimulant type, and viral model.

Table 2 Comparison of Antiviral Recognition Features of RIG-I and MDA5

3 Regulatory Position of IFI35 in the Interferon Response

3.1 Basic features of IFI35

(1) Interferon-induced expression

IFI35 is an interferon-inducible protein and can be upregulated after viral infection, type I interferon stimulation, or inflammatory signaling activation. Its expression changes often appear together with ISG expression profiles, so it can serve as one indicator of interferon-response status.

(2) Feedback regulatory properties

IFI35 is not a typical viral RNA sensor, nor is it a direct antiviral enzyme. It is better understood as a regulatory factor in the interferon response and may participate in modulating RLR-related signaling, inflammatory output, or ISG expression intensity. When studying IFI35, attention should be paid to its effects on IFNB1, ISG15, MX1, OAS1, IFIT1, CXCL10, and other markers.

(3) Cell-context dependence

The role of IFI35 may be affected by cell type, virus type, stimulation intensity, and time window. In some models, IFI35 may function as a negative feedback regulator that limits excessive interferon responses. In other inflammatory or infectious contexts, increased IFI35 expression may accompany stronger immune activation. Therefore, IFI35 upregulation alone should not be used to determine whether it promotes or inhibits antiviral immunity.

3.2 Relationship between IFI35 and RIG-I/MDA5 signaling

(1) Located downstream of recognition signaling

IFI35 is usually not a direct viral RNA recognition receptor. Instead, it is positioned at the level of interferon-induced responses and feedback regulation. After RIG-I or MDA5 activation, IFNB1 and the ISG program are initiated, and IFI35 can enter the regulatory network as one of the induced genes.

(2) Influence on response amplitude

If IFI35 knockdown increases IFNB1, ISG15, IFIT1, MX1, or CXCL10 expression, this suggests that IFI35 may limit the interferon response. If these genes decrease after knockdown, IFI35 may support maintenance of antiviral signaling. Any conclusion must be combined with viral replication readouts and upstream phosphorylation signals.

(3) Association with viral replication outcomes

Whether altered IFI35 expression has antiviral significance should be assessed through viral RNA copy number, viral protein, proportion of infected cells, or viral titer. If IFI35 knockdown changes ISG expression but does not significantly affect viral replication, IFI35 may mainly regulate signaling phenotype rather than act as a decisive antiviral effector.

3.3 IFI35 experimental design

(1) Expression induction experiments

Cells can be stimulated with IFN-β, IFN-α, viral infection, 5′ppp RNA, or Poly(I:C) to assess time-dependent IFI35 expression. Time points such as 2 h, 6 h, 12 h, 24 h, and 48 h are recommended to distinguish early signaling genes, classical ISGs, and late feedback factors.

(2) Knockdown or overexpression experiments

IFI35 functional validation can use siRNA, shRNA, CRISPR, or overexpression systems. Knockdown efficiency should be confirmed at both mRNA and protein levels. Overexpression experiments should avoid expression levels far above physiological induction levels, otherwise nonphysiological protein interactions or stress responses may be introduced.

(3) Pathway-layered detection

When evaluating the effect of IFI35 on antiviral signaling, p-TBK1, p-IRF3, IFNB1, IFNL1, STAT1/STAT2 phosphorylation, ISG expression, and viral replication should be detected in layers. Detecting only a single ISG is insufficient to determine whether IFI35 acts at the RLR recognition layer, interferon production layer, or JAK-STAT response layer.

Table 3 Detection Framework for IFI35 Functional Studies

4 ISG Expression Regulation and Antiviral Effects

4.1 Initiation of ISG expression

(1) Type I interferon pathway

After IFN-β and IFN-α bind to IFNAR1/IFNAR2, JAK1 and TYK2 are activated, subsequently promoting phosphorylation of STAT1 and STAT2. STAT1, STAT2, and IRF9 form the ISGF3 complex and bind ISRE elements, inducing expression of multiple ISGs.

(2) Type III interferon pathway

IFN-λ activates similar JAK-STAT signaling through the IFNLR1 and IL10RB receptor complex and mainly exerts antiviral effects in epithelial barrier tissues. In respiratory, intestinal, and hepatic epithelial models, the relationship between IFN-λ and ISG expression deserves separate analysis.

(3) Direct ISG induction

Some ISGs can be directly induced after RLR activation without complete dependence on secreted interferons, especially early IRF3- or NF-κB-related genes. If direct induction and interferon autocrine amplification need to be distinguished, JAK inhibitors, IFNAR-blocking antibodies, or conditioned-medium transfer experiments can be used.

4.2 Functional grouping of typical ISGs

(1) Viral RNA degradation and replication inhibition

The OAS1, OAS2, OAS3, and RNase L axis promotes viral RNA degradation. PKR/EIF2AK2 inhibits protein translation through eIF2α phosphorylation. The IFIT family can recognize abnormal cap structures or specific viral RNA features and restrict viral replication.

(2) Restriction of viral entry and release

IFITM1, IFITM2, and IFITM3 can affect viral membrane fusion and entry. BST2/Tetherin can restrict the release of some enveloped viral particles. The functions of these ISGs are closely related to viral life-cycle stages, so stage-specific assays for viral entry, replication, and release should be used.

(3) Ubiquitin-like modification and immune regulation

ISG15 can affect host and viral protein functions through ISGylation and is also a common marker of interferon response intensity. MX1/MX2, RSAD2/Viperin, GBP family proteins, and others interfere with viral replication, assembly, or intracellular localization through different mechanisms.

4.3 Interpretation of ISG expression

(1) A single gene is insufficient to represent antiviral status

Upregulation of a single ISG only indicates activation of one pathway node and does not represent the full antiviral state. It is recommended to detect multiple markers such as ISG15, MX1, OAS1, IFIT1, RSAD2, and CXCL10, together with viral replication readouts.

(2) mRNA and protein are not always synchronized

ISG mRNA usually increases earlier, while protein expression is affected by translation, stability, and viral immune evasion. If mRNA is markedly increased but protein remains unchanged, time point, antibody sensitivity, viral inhibition of protein translation, and sample processing should be checked.

(3) Excessive ISG responses may accompany cell injury

High-intensity interferon or dsRNA stimulation can induce cell-cycle arrest, apoptosis, pyroptosis, or metabolic stress. In antiviral experiments, cell viability, death rate, and inflammatory cytokines should be measured in parallel to avoid misinterpreting virus reduction caused by cytotoxicity as a specific antiviral effect.

Table 4 Common ISGs and Antiviral Effect Directions

5 Experimental Models and Method Selection

5.1 Viral infection models

(1) Authentic viral infection

Authentic viral infection can simultaneously present viral entry, replication, immune evasion, and ISG feedback regulation, making it suitable for evaluating the functional significance of the RIG-I/MDA5-IFN-ISG axis. Experiments should be designed according to viral biosafety level, MOI, infection time, and cell type, while measuring both viral replication and host signaling.

(2) Pseudovirus or replicon systems

Pseudovirus and replicon systems are suitable for studying viral entry, replication elements, or reporter signals at lower biosafety risk. These models facilitate high-throughput screening but cannot fully reproduce immune evasion proteins and the complete replication cycle of authentic viruses.

(3) Viral mimic stimulation

5′ppp RNA, Poly(I:C), extracted viral RNA, or synthetic dsRNA can be used to dissect RIG-I/MDA5 signaling. Mimic experiments are more suitable for mechanistic localization, but delivery method, dose, and cytotoxicity must be considered. They cannot fully replace viral infection results.

5.2 Signal detection methods

(1) qPCR

qPCR is suitable for detecting transcriptional changes in DDX58, IFIH1, IFI35, IFNB1, IFNL1, ISG15, MX1, OAS1, IFIT1, RSAD2, CXCL10, and related genes. Time point design should cover early recognition, interferon production, and ISG amplification stages.

(2) Western blot

Western blot can detect protein changes in RIG-I, MDA5, IFI35, p-TBK1, p-IRF3, p-STAT1, ISG15, MX1, and OAS1. For pathway activation studies, phosphorylated proteins and total proteins should be detected together to avoid confusing expression changes with activation changes.

(3) Immunofluorescence and flow cytometry

Immunofluorescence can observe nuclear translocation of IRF3, p65, or STAT1, as well as localization of viral proteins and ISGs. Flow cytometry is suitable for analyzing the proportion of infected cells, ISG-positive cell populations, and cell death. In heterogeneous infection models, single-cell-level detection provides more information than whole-lysate analysis.

5.3 Functional validation methods

(1) Knockdown and knockout

Knockdown or knockout of DDX58, IFIH1, MAVS, IFI35, STAT1, or key ISGs can be used to determine functional positioning. Knockdown experiments should include nontargeting controls, multiple siRNA sequences, and rescue experiments to avoid conclusions affected by off-target effects.

(2) Interferon rescue experiments

If RIG-I or MDA5 deficiency leads to reduced ISG expression, exogenous IFN-β or IFN-λ can be added to determine whether downstream JAK-STAT signaling remains responsive. If exogenous interferon restores ISG expression, the defect is mainly located at the interferon production layer. If it does not, receptor or STAT pathway abnormalities should be considered.

(3) Viral replication readouts

Antiviral function must be validated through viral readouts. Common indicators include viral RNA copy number, viral protein expression, infected-cell proportion, viral titer, plaque formation, and cytopathic effects. ISG upregulation alone cannot prove enhanced antiviral effects.

Table 5 Method Combinations for RIG-I/MDA5-IFI35-ISG Studies

6 Reagent and Material Selection for Interferon and ISG Research

Table 6 Detection and Intervention Products for RIG-I, MDA5, IFI35, TBK1, and IRF3 Signaling Nodes

Table 7 Products Related to JAK-STAT and ISG Expression Detection and Functional Validation

7 Common Questions

7.1 Both RIG-I and MDA5 can induce ISGs. How can their roles be distinguished experimentally?

Their roles should be distinguished by combining stimulants, viral models, and genetic intervention. 5′ppp RNA is more biased toward RIG-I, whereas long dsRNA or high-molecular-weight Poly(I:C) is more commonly used in MDA5-related studies. Final contribution still needs to be validated through DDX58 or IFIH1 knockdown/knockout.

7.2 Is IFI35 an antiviral effector gene or a regulatory factor?

IFI35 is better regarded as an interferon-induced regulatory factor rather than a typical direct antiviral enzyme. Its functional direction should be determined based on IFNB1, ISG expression, and viral replication outcomes after knockdown or overexpression. It should not be defined as antiviral enhancement solely based on expression upregulation.

7.3 Can ISG15, MX1, OAS1, or IFIT1 be selected individually as an ISG marker?

Using only one is not recommended. Different ISGs have different induction kinetics and antiviral mechanisms, so a single marker can easily lead to misinterpretation. At least 3–5 ISGs should be selected, and viral replication and cell viability should be measured simultaneously.

7.4 Does Poly(I:C) stimulation always represent MDA5 activation?

No. The molecular weight, delivery method, and cell type determine the main recognition route of Poly(I:C). Cytosolic delivery of high-molecular-weight Poly(I:C) is more suitable for studying MDA5. When directly added to the culture medium, TLR3 signaling may be the main contributor.

7.5 Does increased RIG-I or MDA5 expression mean that cells are recognizing viruses?

Not necessarily. RIG-I and MDA5 themselves can be induced by interferons, and increased expression may only indicate that cells are in an antiviral alert state. p-TBK1, p-IRF3, IFNB1, MAVS dependence, and viral replication changes should be combined to determine whether true recognition activation exists.

RIG-I and MDA5 are key receptors at the viral RNA recognition layer. IFI35 is located within the interferon-induced and feedback-regulatory network, while ISGs constitute the execution layer of antiviral effects. When studying this pathway, validation should proceed layer by layer from viral recognition, interferon production, JAK-STAT response, and ISG expression to viral replication outcomes, avoiding the direct equation of single-gene expression changes with altered antiviral function.

For more related articles, please see below:

[1] Human α2a, α2b Interferon Assay

[2] Detection of α-, β- and γ-interferon-induced antiviral activity

[3] Type Stratification, Signal Transduction, and Effector Networks of the Interferon (IFN) Pathway