Development, applications, and future prospects of RNA virus reverse genetics technology

1 Introduction

RNA viruses are one of the most important groups of pathogens threatening global public health, livestock industry security, and ecological stability. They include positive-sense single-stranded RNA (+ssRNA) (1), negative-sense single-stranded RNA (-ssRNA) (2), and segmented RNA viruses (3), and are characterized by high mutation rates, rapid evolution, strong adaptability, and frequent cross-species transmission. In recent decades, major epidemics—including COVID-19, seasonal and pandemic influenza, and Newcastle disease—have highlighted the urgent need for deeper mechanistic understanding, effective vaccines, and targeted antivirals. Viral reverse genetics, which enables the rescue of infectious viruses from cloned cDNA, has become the cornerstone technology for dissecting RNA virus life cycles, defining genotype–phenotype relationships, and developing countermeasures (4–6).

Over the past decades, reverse genetics has evolved from simple single-plasmid systems for positive-sense RNA viruses to sophisticated multi-plasmid and modular platforms for negative-sense and segmented RNA viruses. Today, it is widely applied across diverse RNA virus families and serves as an essential tool for viral pathogenesis studies, vaccine design, virus–host interaction analysis, immune evasion research, and high-throughput antiviral screening. Modern systems have been optimized into fully plasmid-based workflows, eliminating the need for helper viruses or in vitro transcription, and significantly improving rescue efficiency and reproducibility. Key innovations include dual-promoter strategies, ribozyme cleavage tools, BAC, infectious subgenomic amplicons (ISA), and yeast homologous recombination, which have greatly expanded the scope of engineerable RNA viruses (7–11).

Despite tremendous advances over recent decades, this field still faces multiple unresolved bottlenecks. First, virus-type-specific technical barriers exist: +ssRNA viruses can be rescued with relative ease, whereas non-segmented and segmented -ssRNA viruses demand complex multi-plasmid co-transfection and ribonucleoprotein (RNP) reconstitution, resulting in poor rescue efficiency (12). Second, genome instability and toxicity are major bottlenecks for large-genome RNA viruses such as coronaviruses, where full-length cDNA clones frequently undergo mutation, deletion, or recombination during bacterial propagation (13). Third, biosafety risks are a critical concern, as reverse genetics enables reconstruction of highly pathogenic viruses, raising risks of laboratory escape and misuse (14). Fourth, the absence of universal, standardized reverse genetics platforms impedes rapid countermeasures against newly emerging viruses (15). Notable highlights include continuous technical innovations overcoming key bottlenecks for large-genome and highly pathogenic viruses, broad translational applications in COVID-19 and influenza vaccine development, and accelerated interdisciplinary integration with CRISPR/Cas editing, artificial intelligence, and big data, driving precision virology and rational antiviral design (16).

Engineered RNA viruses are categorized into three major groups based on genome structure and replication mechanisms, all amenable to reverse genetic manipulation (17, 18). The +ssRNA viruses, such as coronaviruses, have genomic RNA that functions directly as mRNA, supporting independent replication without additional viral proteins (4). The -ssRNA viruses, exemplified by Newcastle disease virus (NDV), have non-infectious genomes that require viral nucleoprotein and polymerase complexes for replication and transcription (19). Segmented negative-sense single-stranded RNA viruses, represented by influenza A virus, possess genomes divided into multiple independent segments, each encoding specific viral proteins and requiring coordinated expression for virion assembly (20).

This review selects SARS-CoV-2 (+ssRNA), NDV (non-segmented -ssRNA), and influenza A virus (segmented -ssRNA) as representative models for the following rationales (21). First, comprehensive genome representativeness: these three viruses fully cover the three major genome types of RNA viruses, and their reverse genetics systems are the most mature and widely adopted for each category, enabling a systematic overview of technical principles and optimization strategies (22). Second, typical research challenges: SARS-CoV-2 exemplifies large-genome instability and cloning difficulties; NDV represents non-segmented -ssRNA viruses requiring multi-plasmid RNP reconstitution; influenza A virus embodies the complexity of segmented genome assembly and balanced expression, collectively reflecting core technical bottlenecks in the field (23). Third, high translational value: SARS-CoV-2 is a pandemic pathogen driving global vaccine and antiviral development; NDV is a well-established vaccine vector and oncolytic virus platform; influenza A virus is a model for vaccine strain engineering and pandemic preparedness, spanning basic research and clinical translation (24). Fourth, broad reference significance: research on these three viruses provides universal insights for other RNA viruses of the same type, supporting cross-family reverse genetics strategy development and rapid responses to emerging pathogens (25).

Notably, reverse genetics technology has also been widely applied to numerous other high-risk pathogens with prominent public health significance, among which Nipah virus and Ebola virus have been thoroughly characterized in abundant published studies (25, 26). However, both pathogens belong to non-segmented negative-sense RNA viruses, which cannot achieve full coverage of all three major RNA virus genome categories as the three selected prototype viruses do. In addition, the rescue and manipulation of live Nipah and Ebola viruses require BSL-4 high-containment laboratories, which brings unique experimental limitations and biosafety barriers (27, 28). For the above reasons, they are not included as core research models in this review and will only be briefly discussed in the section focusing on emerging and re-emerging viruses.

Existing review articles generally focus on a single RNA virus genus and lack horizontal cross-comparisons among +ssRNA, non-segmented −ssRNA and segmented −ssRNA reverse genetics platforms. From the comprehensive comparison in this review, we argue that few prior works systematically summarize shared technical bottlenecks and tailored optimization strategies applicable to all three major genome types, which cannot provide universal experimental guidance for constructing rescue systems for uncharacterized emerging RNA viruses. Moreover, previous summaries only briefly mention Nipah and Ebola viruses without linking their technical frameworks to the classic NDV rescue system, forming a critical logical gap in the field.

Based on the above background, this review systematically summarizes the development history, core technical components, optimization strategies, and latest research progress of RNA virus reverse genetics technology, focusing on the three representative viruses of SARS-CoV-2, NDV, and influenza A virus. We analyze the key challenges and corresponding solutions in the construction of reverse genetics systems for different types of RNA viruses, and discuss their wide applications in vaccine development, pathogenic mechanism research, and antiviral drug screening. Finally, we look forward to the future development directions of this technology, including technical modular standardization, cross-disciplinary integration, biosafety improvement, and rapid response to emerging viruses. This review aims to provide a comprehensive, systematic, and in-depth technical reference for basic research and prevention–control applications of RNA viruses, and promote the further development and application of reverse genetics technology in the field of virology.

2 Methods

This manuscript was designed as a systematic narrative review focusing on the developmental history, technical framework, application scenarios, existing bottlenecks and future research directions of RNA virus reverse genetics technology; this review only summarizes published technical optimization strategies, core vector components and translational research outcomes, and does not involve human participants, animal infection experiments or clinical intervention trials. Qualitative narrative synthesis was adopted for subsequent data analysis, through which different reverse genetics technical platforms were classified, multi-dimensional application scenarios were summarized, consensus research conclusions were integrated, current technical limitations and prospective research trends were analyzed, and quantitative meta-analysis was not conducted in this review; a PRISMA flow diagram illustrating the full literature screening and study selection process is provided as Figure 1, with detailed screening statistics as follows: Literature retrieval was conducted in PubMed, Web of Science, Scopus and Embase. After strict screening by title, abstract and full text, 410 peer-reviewed articles were included for qualitative narrative synthesis without meta-analysis; the PRISMA screening flowchart is shown in Figure 1.

3 Basic principle & technical processes of RNA virus reverse genetics

3.1 Core principles and technical processes

Viral reverse genetics is a core molecular biology technique, whose basic principle is to reconstruct infectious viral particles with specific genotypes and phenotypes through in vitro manipulation of viral genomes (29). Critically, forward genetics relies on untargeted random mutation screening, which fails to establish one-to-one correlations between single genetic loci and viral phenotypes; reverse genetics enables rational, site-directed genome editing, making it an irreplaceable technical platform for precise virological research that random mutagenesis cannot replicate.

The development of this technology can be traced back to the mid-1970s, when the first successful construction of infectious clones of positive-strand RNA (poliovirus) viruses marked a major breakthrough in the methodology of virological research ⁠ (29). Since then, with the advancement of recombinant DNA technology, reverse genetics systems have been widely applied in the research of various RNA viruses ⁠ (30, 31).

The establishment of a reverse genetics system follows a universal basic workflow. First, viral RNA genome is converted into cDNA by reverse transcription as the prerequisite for DNA-level manipulation. Comment: RNA degradation and low reverse transcription efficiency for high-variability viral genomes are primary technical barriers at this step (32, 33). The obtained cDNA is then cloned into an appropriate vector to form an infectious clone carrying complete viral genome and essential regulatory elements. Comment: Large viral cDNA fragments (e.g., ~30 kb SARS-CoV-2 genome) often trigger bacterial toxicity and genetic deletion during cloning (34). After genetic modification (mutation, deletion, reporter gene insertion, etc.), the recombinant construct is introduced into host cells to initiate viral replication and rescue infectious virions. Comment: Multi-plasmid co-transfection required for -ssRNA viruses drastically reduces the proportion of fully transfected cells and rescue titers. Notably, non-segmented negative-strand RNA viruses require co-delivery of auxiliary plasmids encoding NP/P/L to reconstitute functional RNP complexes for successful rescue. Comment: Unbalanced expression of RNP components further suppresses rescue efficiency of recombinant viruses (21, 22).

Conventional forward genetics relies on random mutagenesis and cannot precisely correlate single genomic mutations with viral phenotypic changes. Based on systematic sorting of functional research cases, we hold the view that the fundamental advantage of reverse genetics is enabling rational, targeted viral engineering, rather than merely dissecting individual viral gene functions. It serves as the exclusive technical platform supporting rational live attenuated vaccine design and targeted antiviral target verification, which explains its complete replacement of traditional virological phenotypic screening approaches.

In summary, the standardized reverse genetics workflow is shared across all RNA virus groups, yet each step faces type-specific technical bottlenecks. Unlike blind forward genetic screening, the genotype-first editing logic of reverse genetics enables targeted exploration of virulence loci, immune escape sites and drug targets, laying the foundational technical premise for subsequent vaccine and antiviral development.

3.2 Genome-type-specific strategies and unique challenges for RNA virus reverse genetics

RNA viruses encompass three major genome types (positive-sense single-stranded, negative-sense single-stranded), which require highly tailored reverse genetics strategies rather than a universal protocol (Figures 2–4). The type-specific cloning strategies, preferred promoters and core technical bottlenecks of the three RNA virus categories are systematically summarized in Table 1 for intuitive cross-comparison.

Table 1. Genome-type-specific reverse genetics strategies and challenges for RNA viruses.

As summarized in Table 1, the three categories of RNA viruses face distinct experimental obstacles. This review concludes that +ssRNA coronaviruses feature straightforward single-plasmid transfection but suffer severe cytotoxicity and instability of ~30 kb full-length cDNA; non-segmented −ssRNA viruses such as NDV require cumbersome multi-plasmid RNP co-transfection and generate low rescued viral titers; segmented influenza viruses struggle to balance synchronized expression of eight independent genomic segments. No single universal rescue protocol is adaptable to all RNA virus clades, which greatly restricts the widespread application of reverse genetics technology.

3.2.1 Positive-sense single-stranded RNA viruses

Representative viruses include SARS-CoV-2. The viral genomic RNA itself acts as mRNA and is directly translatable in host cells. The core strategy relies on direct construction of full-length infectious cDNA clones, driven by eukaryotic promoters (CMV/Pol II) or T7 promoters for in vitro transcription. Transfection of the full-length cDNA construct alone is sufficient to initiate viral replication and rescue infectious virions (35). This single-plasmid rescue scheme greatly simplifies the construction of SARS-CoV-2 infectious clones and accelerates the rapid preparation of attenuated vaccine seed strains.

Overall, the single-plasmid rescue workflow of +ssRNA viruses simplifies experimental operations severely, yet inherent bacterial instability of oversized full-length cDNA clones remains an insurmountable intrinsic limitation of this system.

3.2.2 Negative-sense single-stranded RNA viruses

Reverse genetics systems for non-segmented negative-sense single-stranded RNA viruses are typically represented by NDV. This system requires the construction of a full-length antigenomic cDNA plasmid complying with the “rule of six”, paired with three helper plasmids expressing the core NP, P, and L proteins. Relying on T7 RNA polymerase, viral genome transcription and replication take place in the cytoplasm, and infectious recombinant viruses are ultimately rescued via co-transfection of these plasmids into host cells (36). The RNP co-transfection mode established by this strategy provides a standardized operation template for constructing NDV live vector vaccines and oncolytic viruses.

Although the T7-dependent four-plasmid RNP reconstitution system achieves stable rescue of NDV, low co-transfection efficiency inevitably lowers recombinant virus yields, restricting its application in high-throughput screening.

3.2.3 Segmented RNA virus

As a typical segmented negative-sense single-stranded RNA virus, influenza virus uses a reverse genetics system consisting of eight viral RNA segment expression plasmids combined with four helper plasmids encoding polymerases and nucleoprotein. With the aid of dual Pol I/Pol II promoters, intact viral RNAs (vRNAs) are transcribed in the nucleus and assembled into infectious virions (37). The segmented multi-plasmid cloning framework supports rapid reassortment of HA/NA antigen genes, becoming the core production technology for seasonal influenza vaccine seed viruses.

The dual Pol I/Pol II eight-plasmid system realizes synchronized transcription of eight viral segments, but imbalanced expression between segments frequently causes incomplete virion assembly and low rescue success rates.

3.3 Non-infectious reverse genetics systems: pseudoviruses, minigenomes and subgenomic replicons

Although full-length infectious cDNA clones support the rescue of intact live viruses and recapitulate the complete viral life cycle, they impose prominent limitations for highly pathogenic RNA viruses, including strict BSL-3/BSL-4 laboratory requirements, elevated biosafety risks, high experimental costs, and incompatibility with ultra-high-throughput screening. A suite of replication-deficient, non-infectious research platforms have been developed based on reverse genetic engineering principles. These systems only retain specific viral functional elements coupled with reporter genes such as luciferase and EGFP, enabling mechanistic research, neutralizing antibody evaluation and high-throughput primary screening of antiviral compounds under BSL-2 conditions. They serve as indispensable complementary tools to full-length infectious clones, mainly categorized into five types: pseudoviruses, minigenomes, subgenomic replicons, ISA-derived replicons and episomal replicons. Their construction principles, representative applications and research progress are summarized as follows.

3.3.1 Pseudovirus systems

Pseudoviruses are replication-deficient viral particles whose core genomic backbone is fully replaced by reporter genes (e.g., luciferase, EGFP), while the envelope proteins of target viruses are displayed on the particle surface. Such particles can only mediate viral attachment and membrane fusion to enter host cells, but fail to support intracellular genome replication and progeny virion assembly and release. No risks of laboratory leakage or cross-species transmission exist, and all experimental manipulations can be conducted in BSL-2 laboratories. Pseudovirus systems based on HIV/VSV backbones are mainly applied to SARS-CoV-2 for neutralizing antibody and entry inhibitor screening, and also adapted to multiple flaviviruses for low-biosafety research (38–40).

3.3.2 Minigenome systems

Minigenomes retain only cis-regulatory sequences required for viral replication and transcription at both termini of the viral genome, while all viral structural and non-structural protein coding regions are substituted with luciferase or EGFP reporter cassettes. Co-transfection of helper plasmids expressing NP, P and L proteins reconstitutes functional viral ribonucleoprotein (RNP) complexes to drive RNA replication and reporter transcription in host cells without generating fully infectious virions. Minigenomes represent core tools for dissecting replication mechanisms of negative-sense RNA viruses.

Minigenome platforms have been established for segmented and non-segmented negative-sense RNA viruses. For influenza A virus, a minigenome system driven by feline RNA polymerase I enables precise quantification of RdRp activity across distinct cell types, providing critical evidence for interspecies transmission research (41). For NDV, a minigenome rescue system based on the attenuated D58 strain retains only the minimal genomic cis-elements; co-expression of NP, P and L helper plasmids drives robust reporter expression and lays a foundation for full-length infectious clone construction (42). For SARS-CoV-2, two distinct minigenome workflows have been developed: in vitro transcribed viral RNA or multi-plasmid co-transfection to supply complete nsp repertoires, with reporter luminescence directly reflecting viral replication levels for polymerase inhibitor primary screening under BSL-2 conditions (43).

3.3.3 Subgenomic replicons

Subgenomic replicons preserve full coding sequences of viral replication-associated non-structural proteins, while structural genes (S, M, E) are deleted and replaced with luciferase or GFP reporter cassettes. They autonomously replicate in the cytoplasm but cannot assemble mature infectious virions due to missing structural components, permitting long-term serial passaging for sustained monitoring of viral replication and evolutionary dynamics of drug resistance mutations.

Multiple modular SARS-CoV-2 subgenomic replicons have been reported. S-deleted stable replicons support high-throughput screening of antiviral compounds under BSL-2 containment (44). A toolkit carrying nsp1 loss-of-function mutations (K164A and H165A) allows flexible swapping of diverse reporter and selection markers (45). Dual-reporter replicons equipped with Rluc-GFP and Fluc-RFP cassettes simultaneously quantify viral replication activity and are specialized for screening RdRp inhibitors (46).

3.3.4 ISA-derived replicons

Infectious Subgenomic Amplicon (ISA) technology is an innovative reverse genetic strategy that bypasses full-length bacterial plasmid cloning by assembling intact viral nucleic acids via intracellular homologous recombination of overlapping DNA fragments. Deletion of structural gene segments yields ISA-derived replicons devoid of infectivity, effectively resolving major bottlenecks of full-length cDNA instability and bacterial cytotoxicity for large-genome viruses.

ISA workflows have been deployed for flaviviruses including dengue, Japanese encephalitis and West Nile viruses, circumventing mutagenesis and deletion artifacts observed during bacterial propagation of full-length viral plasmids (47, 48). Notably, ISA subgenomic fragments can be directly delivered into live animals to support viral replicative element expression, establishing low-risk preclinical platforms for antivirals and vaccine candidates (49). Nevertheless, cytotoxicity of certain ISA fragments during bacterial amplification reduces fragment yields, constituting a key drawback of this approach.

3.3.5 Episomal replicons

Episomal replicons are autonomous DNA or RNA molecules that replicate independently of host chromosomes without genomic integration, sustaining persistent viral protein or reporter expression. Alphavirus episomal replicons serve as classic heterologous expression backbones with structural genes replaced by foreign sequences (50). Although most established SARS-CoV-2 replicons replicate in the cytoplasm rather than the nucleus, they functionally recapitulate core features of episomal systems: autonomous viral RNA replication in the absence of infectious virion production, supporting long-term monitoring of replicative activity for durable antiviral evaluation and viral non-structural protein functional studies (44, 46).

3.3.6 Standard three-tier antiviral screening pipeline integrating non-infectious surrogates and full-length infectious clones

Combined application of the five non-infectious platforms forms a standardized tiered evaluation workflow for antiviral discovery: (1) Primary high-throughput screening via pseudovirus systems under BSL-2 to rapidly filter entry inhibitors and neutralizing antibodies; (2) Secondary mechanistic validation utilizing minigenomes and subgenomic replicons to quantify polymerase inhibition and monitor the emergence of drug resistance; (3) Definitive functional verification with full-length infectious reporter viruses under matching BSL-3/BSL-4 containment to measure comprehensive antiviral potency, viral pathogenicity and in vivo protective effects. Representative clinical drugs follow this pipeline during preclinical development: pseudovirus platforms completed primary screening of neutralizing antibodies against SARS-CoV-2; minigenome and replicon systems validated the RdRp inhibitory activity of remdesivir and molnupiravir; influenza minigenome assays screened the polymerase inhibitor baloxavir marboxil, while pseudovirus platforms evaluated the neuraminidase inhibitor oseltamivir at early screening stages.

To systematically visualize the differentiation of mainstream reverse genetics experimental platforms, we summarize core performance indicators across full-length infectious clones, pseudoviruses, minigenomes and subgenomic replicons in newly supplemented Table 2, with comparison dimensions including biosafety restriction, in vivo experimental feasibility, compound screening throughput, research cost and evaluable viral functional modules.

Table 2. Comparative profiles of five non-infectious reverse genetics surrogate systems.

None of the five surrogate systems support in vivo pathogenicity or vaccine efficacy tests; pseudoviruses cannot assess replication; minigenomes/replicons fail to screen assembly-targeted compounds.

As clearly listed in Table 2, each surrogate system bears exclusive experimental limitations that cannot be ignored: pseudoviruses fail to assess viral replication, assembly and in vivo infection; minigenomes lack entry and virion maturation readouts; subgenomic replicons cannot screen assembly or release-targeted antivirals. Researchers must select matching platforms according to specific research endpoints rather than relying solely on low-biosafety surrogate tools.

From the comprehensive comparison in this review, we propose a core critical perspective: all non-infectious surrogate systems only recapitulate partial fragments of the complete viral life cycle, leading to inevitable phenotypic deviation between surrogate readouts and wild-type viral infection. None of these low-biosafety platforms can fully substitute full-length infectious cDNA clones for in vivo pathogenicity testing, vaccine efficacy assessment or multi-step viral replication mechanism research.

4 Core components and optimization strategies of reverse genetics technology

Early viral reverse genetics systems typically relied on relatively complex operational procedures. For example, helper viruses (viruses used to provide replication proteins or polymerases for the target virus) were required to provide polymerases necessary for viral replication, or in vitro RNA synthesis followed by transfection into cells was needed to initiate viral replication. These methods were often inefficient and cumbersome, limiting the ability to conduct large-scale or high-throughput studies on mes (17, 51).

To overcome these limitations, modern reverse genetics systems have been generally optimized into a fully plasmid-based model. In this model, researchers co-transfect a set of carefully designed plasmids into host cells. These plasmids act synergistically to not only provide cDNA copies of the viral genome but also supply all viral proteins essential for replication, transcription, and translation (12). This approach eliminates the need for in vitro transcription steps, simplifies the experimental workflow, and significantly improves the efficiency of virus rescue.

In full-plasmid reverse genetics systems, the selection and fine-tuning of transcriptional elements are critical, as they directly affect the efficiency and accuracy of viral RNA synthesis, which in turn determines the production of infectious viral particles.

To systematically improve the rescue efficiency and genome stability of the three categories of RNA virus reverse genetics systems summarized in Section 3, researchers have carried out targeted optimization from six core dimensions: vector backbone selection, promoter transcriptional element matching, ribozyme terminal modification, multi-plasmid system simplification, large viral genome assembly technology and exogenous gene insertion engineering. The following subsections discuss each optimization module in sequence, and compare the applicable scenarios of different tools combined with SARS-CoV-2, NDV and influenza A virus models.

4.1 Vector systems

Vectors are key tools for carrying viral cDNA and expressing auxiliary proteins. For viral reverse genetics, the selection of vectors needs to consider their stability in bacteria and expression efficiency in eukaryotic cells. Commonly used vectors in viral reverse genetics systems include pcDNA3.1 and p-Bluescript (52, 53).

pcDNA3.1 is a widely used mammalian expression vector. Belonging to the pcDNA series, it is designed to achieve high-efficiency expression of inserted genes in mammalian cells. In the research aimed at constructing a universal reverse genetics vector for negative-strand RNA viruses, pcDNA3.1 is usually used as the expression vector for the auxiliary plasmid, eliminating the need for in vitro transcription of the auxiliary protein and thereby enhancing the virus rescue efficiency. As a low-copy cloning vector, pBluescript stabilizes large viral genomes against mutations during bacterial amplification, rendering it ideal for full-length viral cDNA cloning. Moreover, it contains a T7 RNA polymerase promoter, enabling the in vitro transcription of viral RNA (Table 3).

Table 3. Application of vector systems in RNA virus reverse genetics.

The reverse genetics of the influenza virus has been updated and refined multiple times. The latest vector system mainly consists of pDP2001/pDZ/pComb8 (vRNA) + optimized pcDNA3.1/pCAGGS (helper protein); multi-fragment combined vectors (2–4 plasmids) are the most mainstream optimization direction in recent years (60, 61). Rational selection of low-copy vectors and mammalian expression backbones effectively solves the problem of large viral genome instability and improves the repeatability of virus rescue experiments.

Low-copy pBluescript guarantees stable propagation of large viral cDNA fragments, while pcDNA3.1 supports robust auxiliary protein expression in mammalian cells. We comment that neither backbone can simultaneously satisfy the dual demands of large-genome stability and high-level eukaryotic expression. Commercially available vectors lack pre-designed multi-cloning sites optimized for viral genome insertion, forcing researchers to conduct tedious self-modification and prolonging experimental cycles. Chimeric shuttle vectors combining the merits of low-copy backbones and strong eukaryotic expression cassettes will become a vital optimization direction in the future.

4.2 Promoter systems

In reverse genetics systems, the selection and optimization of transcriptional elements, particularly promoters, are critical steps to ensure the efficient and accurate synthesis of viral RNA in host cells. These systems generate viruses through genetic manipulation and are widely applied in vaccine development, research on viral pathogenic mechanisms, and antiviral drug screening (62, 63). Depending on the specific characteristics of viruses and expression requirements, researchers select and modify various promoters to drive the expression of viral genomes or auxiliary proteins.

4.2.1 T7 RNA polymerase promoters

The T7 RNA polymerase promoter is a robust promoter derived from bacteriophages, which can efficiently initiate transcription in prokaryotic systems (64, 65). To exert its function in eukaryotic cells, the T7 promoter usually requires the exogenous expression of T7 RNA polymerase (66–69). This mechanism enables its extensive application in in vitro transcription and viral reverse genetics systems within eukaryotic cells. For instance, in the genetic construction of the NDV S519G mutant strain, the T7 promoter and T7 terminator were also utilized to drive the transcription of viral genes (70, 71). Although the mainstream approach of the influenza virus reverse genetics system is based on RNA polymerase I (Pol I), some studies have developed reverse genetics systems using T7 RNA polymerase (66). This system achieves the transcription of influenza virus genomic RNA through a vector containing the T7 RNA polymerase promoter, hepatitis delta virus ribozyme sequence, and T7 RNA polymerase terminator sequence (32). Cytoplasmic transcription driven by T7 polymerase is the core technical basis for the rescue of NDV and other non-segmented negative-sense RNA viruses.

4.2.2 Pol I promoters

RNA polymerase I is primarily responsible for the synthesis of ribosomal RNA (rRNA) (72, 73). Its products are uncapped and localized within the nucleus. The use of Pol I promoters for viral RNA synthesis allows for the efficient production of uncapped viral genomic RNA in the nucleus. As a segmented negative-strand RNA virus, the replication of influenza virus genomic RNA and the synthesis of viral mRNA both occur in the nucleus; thus, its reverse genetics system extensively utilizes Pol I promoters to guide the synthesis of viral genomic RNA (74, 75). Studies have shown that the mechanisms of Pol I recognition and promoter melting differ from those of Pol II and Pol III (72). Nuclear uncapped transcription mediated by Pol I realizes the authentic replication of influenza vRNA, which is irreplaceable for segmented virus reverse genetics.

4.2.3 Pol II promoters

RNA polymerase II (Pol II) is mainly responsible for the synthesis of messenger RNA (mRNA) (76–78). Its products are typically capped and transported out of the nucleus for translation. In many reverse genetics systems, Pol II promoters are used to express viral genomic mRNA or auxiliary proteins. For example, in the plasmid expression system of influenza virus, Pol II promoters are employed to transcribe mRNA encoding viral proteins (63). Furthermore, the interaction between enhancers and promoters, along with the dynamics of Pol II, jointly regulates gene expression (79, 80). Enhancers can act on promoters over long distances, bringing them close together through mechanisms such as DNA looping, and collaboratively recruit Pol II and other transcription factors to enhance transcriptional efficiency. The pausing of Pol II in the proximal promoter region also plays an important role in gene expression regulation, and the release of paused Pol II can trigger rapid gene activation (81).

The cytomegalovirus (CMV) promoter is a strong constitutive promoter widely used in eukaryotic expression systems, capable of driving high-level gene expression in various cell types (82). Researchers inserted the receptor-binding domain (RBD) of the SARS-CoV-2 spike protein into the hemagglutinin (HA) molecule or the open reading frame of truncated non-structural protein 1 (NS1) of the live attenuated influenza vaccine virus (83). In this context, the CMV promoter may be utilized to drive the expression of these chimeric antigens during the initial cloning and validation stages, ensuring high-level antigen production and facilitating immunogenicity evaluation. In the reverse genetics platform for SARS-CoV-2, the CMV promoter is integrated into plasmids to drive the in vitro assembly of infectious full-length cDNA, which is critical for the genetic manipulation and functional research of the virus (84). Constitutive high expression without exogenous polymerase simplifies the DNA-launched rescue of coronaviruses and other positive-sense RNA viruses.

4.2.4 Dual-promoter systems

The dual-promoter system introduces two different promoters into the same expression vector, jointly driving the transcription of the viral genome RNA, thereby achieving more precise and efficient regulation. The Pol I–Pol II dual-promoter strategy is widely applied in the reverse genetic systems of influenza virus, SARS-CoV-2 and NDV (9, 85, 86). The Pol I promoter transcribes viral genomic RNAs without 5′ caps and 3′ poly(A) tails, which perfectly recapitulate the structural features of native viral RNAs. According to the characteristics of different viruses, researchers can optimize the species specificity of Pol I promoters, and further modify the system by combining the T7 promoter or BAC vectors. This strategy resolves key bottlenecks, including the instability of full-length cDNA cloning in coronaviruses and the dependence of traditional NDV rescue systems on T7 polymerase-stably expressing cell lines. Meanwhile, it greatly improves viral rescue efficiency, vector stability and experimental universality, and allows related experiments to be conducted at a lower biosafety level (87, 88). Collectively, this system facilitates the research on viral pathogenic mechanisms, screening of antiviral drugs, and the development of novel vaccines.

Key differences between T7 and CMV promoters are summarized in Table 4, with standardized selection criteria for different viral groups. According to the transcriptional characteristics summarized in Table 4, this review establishes standardized selection principles for promoter matching: cytoplasmic non-segmented −ssRNA viruses should adopt T7 promoter paired with dual ribozymes; nuclear-replicating segmented influenza viruses rely on Pol I/Pol II dual-promoter cassettes; cytoplasmic +ssRNA coronaviruses directly utilize single CMV Pol II promoters. Although dual-promoter frameworks possess the broadest compatibility, species-specific Pol I sequences limit cross-species viral engineering, which remains an unresolvable technical bottleneck at present.

Table 4. Comparison between T7 and CMV (Pol II) promoters and criteria for infectious clone construction.

In summary, the T7 promoter is widely used for cytoplasmic transcription of negative-sense RNA viruses; however, its transcripts may carry non-templated nucleotides at 5′/3′ ends and generate immunogenic dsRNA byproducts via 3′ loop-back. Terminal authenticity can be improved by combining T7 transcription with ribozyme cleavage strategies. The CMV promoter is more convenient for positive-sense RNA viruses because it uses host polymerases and supports straightforward DNA-launched virus rescue. The choice between T7 and CMV should be determined by viral genome type, replication compartment, and tolerance to transcript modifications (64, 65, 84).

Each promoter system exhibits distinct characteristics and applicable scenarios. The T7 promoter enables highly efficient transcription but requires the supply of exogenous T7 RNA polymerase. The Pol I promoter is particularly suitable for viruses that replicate in the nucleus, such as influenza virus, as it drives the synthesis of uncapped viral genomic RNA. The Pol II/CMV promoter supports strong and constitutive gene expression in eukaryotic cells, yet generates capped transcripts that may interfere with the rescue of certain RNA viruses. The dual-promoter system combines the strengths of different promoters, achieving a balanced improvement in both transcriptional efficiency and precision. Pol I/Pol II dual-promoter integrates the advantages of nuclear viral RNA transcription and protein expression, and has become the mainstream universal transformation scheme for RNA viruses.

4.3 Ribozyme tools

As a class of catalytically active RNA molecules, ribozymes can precisely cleave RNA strands and play a pivotal role in reverse genetics systems—particularly in ensuring that viral RNA transcripts possess accurate 3’ terminal length and sequence, which is crucial for the stability and replication of the viral genome (11). Reverse genetics systems are powerful molecular genetic tools used to investigate viral gene function, disease pathogenesis, and virus-host interactions, while also facilitating vaccine development (18). The application of ribozymes has significantly improved the efficiency and precision of constructing reverse genetics systems. Hammerhead and HDV ribozymes precisely trim the two ends of transcribed viral RNA, eliminating the interference of abnormal terminal sequences on viral replication and packaging efficiency.

4.3.1 Hammerhead ribozyme and hepatitis delta virus ribozyme

The HHR represents a class of compact, structurally concise and modular catalytic RNA elements. It consists of a conserved core domain and three flanking helical structures, two of which are responsible for efficient self-cleavage under physiological conditions. This ribozyme catalyzes the hydrolysis of phosphodiester bonds, yielding 2’,3’-cyclic phosphate and 5’-hydroxyl termini with considerable in vitro catalytic activity. In viral reverse genetics systems, HHR is routinely engineered at the 5′ terminus of viral cDNA. Following transcription, site-specific self-cleavage mediated by HHR precisely shapes the authentic 5′ end of viral RNA, a prerequisite for the generation of replication-competent and infectious viral particles (89–92).

The HDV ribozyme is another well-characterized self-cleaving ribozyme, which is endogenously encoded by both genomic and antigenomic RNA of the HDV. Distinct from HHR, the HDV ribozyme is commonly fused to the 3′ end of viral cDNA constructs. It mediates post-transcriptional self-cleavage to eliminate exogenous non-viral sequences, thereby guaranteeing the structural accuracy of the 3′ terminus of viral genomic RNA. Aberrant or elongated 3′ terminal sequences would severely compromise viral packaging efficiency and subsequent infection progression (66, 93, 94). Therefore, the HDV ribozyme is indispensable for sustaining viral replication competence and infectivity (Table 5).

Table 5. Application of hammerhead and HDV ribozymes in RNA virus reverse genetics systems.

T7 transcription produces extra terminal nucleotides and immunogenic dsRNA byproducts, limiting rescue efficiency; hammerhead and HDV ribozymes can remove abnormal RNA ends to solve this defect (100–102).

Hammerhead and HDV ribozymes precisely trim aberrant terminal nucleotides generated by T7 in vitro transcription. Our perspective is that ribozyme elements represent low-cost, modular modifications that ought to be standardized accessories for all T7-driven antigenomic plasmids for negative-sense RNA viruses. However, self-cleavage efficiency varies drastically across different cell lines, and residual aberrant dsRNA intermediates induced by T7 transcription cannot be fully eliminated by ribozyme cleavage, which moderately suppresses final virus rescue yields.

4.4 System simplification and plasmid integration

The core of viral reverse genetics technology lies in the efficient and stable recovery of infectious viruses from cloned cDNA constructs. With the advancement of molecular biology techniques, reverse genetics systems have undergone continuous optimization to enhance their efficiency, modularity, and safety.

Early reverse genetics systems were typically complex and inefficient, often relying on helper viruses to provide polymerases or transfection with in vitro-transcribed RNA (103). Modern reverse genetics systems have been generally optimized into fully plasmid-based models, wherein a set of plasmids are co-transfected to provide the viral genome (or its cDNA) and all proteins required for replication (104).

Many RNA viruses, particularly negative-strand RNA viruses, have non-infectious genomes that require virus-encoded RNA-dependent RNA polymerase (RdRp) and its auxiliary proteins to initiate replication and transcription (18). For instance, the genome of NDV is a negative-sense single-stranded RNA, and its replication and transcription processes rely on a polymerase complex consisting of the nucleoprotein (NP), phosphoprotein (P), and large protein (L). In early reverse genetics systems for such viruses, the full-length viral genomic cDNA was typically cloned into one plasmid, whereas the genes encoding the NP, P, and L proteins were separately inserted into distinct expression plasmids (12). This necessitated co-transfection of at least four plasmids (one genomic plasmid and three auxiliary protein expression plasmids) to recover infectious viruses. Such a multi-plasmid co-transfection strategy has several drawbacks:

• Reduced transfection efficiency: The efficiency of cells simultaneously uptaking and expressing multiple exogenous plasmids is far lower than that of a single or a few plasmids. As the number of plasmids increases, the proportion of cells successfully co-transfected with all essential plasmids declines sharply, leading to low viral rescue efficiency (105, 106).
• Increased operational complexity: The preparation, quantification, and combination of each plasmid require substantial time and effort, increasing experimental tediousness (105, 106).
• System instability: A higher number of plasmids may cause imbalanced expression among different components, impairing the correct assembly of the viral replication complex and thereby reducing the success rate and reproducibility of viral rescue (105, 106).

To overcome these challenges, researchers have integrated multiple genes into a single plasmid. One study demonstrated that integrating the genes encoding NDV auxiliary proteins NP, P, and L into a single plasmid yielded a two-plasmid system (one for the viral genome and one for all auxiliary proteins). Compared with the traditional four-plasmid system, this two-plasmid system improved viral rescue efficiency by approximately 10-fold (1 log₁₀) and enabled faster production of infectious viruses (105, 107, 108). In contrast, the eight-plasmid system for influenza virus reverse genetics possesses considerable optimization potential. For example, several studies have employed the Pol I promoter to drive vRNA synthesis, while the Pol II promoter on the same vector concurrently mediates the transcription of corresponding mRNA, thereby realizing bidirectional expression (109).

Integrating NP, P and L auxiliary genes into one single plasmid elevates NDV rescue efficiency by approximately 10-fold compared with the classic four-plasmid system. Through balanced evaluation, we argue that there exists an unavoidable trade-off between plasmid quantity and plasmid genetic stability. Excess gene fusion within one backbone raises the risk of fragment deletion during bacterial amplification; therefore, unlimited plasmid integration cannot achieve an absolute optimal rescue system, and researchers must balance plasmid number according to viral genome size.

4.5 Large genome solutions

Initial reverse genetics systems relied on constructing full-length infectious clones of viral genomes, which posed significant challenges in terms of construction and stability for viruses with large genomes or genetic toxicity (SARS-CoV-2) (110). To address these issues, various optimization strategies have been developed.

4.5.1 ISA approach

ISA technology holds significant and diverse application value in the research of large-genome viruses such as SARS-CoV-2 (7). This technology can provide a reliable technical platform for the development of live attenuated vaccines, enabling the rapid generation of safe and highly immunogenic vaccine candidates through genetic modification (13). It can also construct fluorescently labeled reporter viruses to support high-throughput antiviral drug screening (7). Meanwhile, it allows for the rapid assembly of viral genomes and virus rescue, and can perform precise site-directed mutagenesis on target genes, facilitating the exploration of the functional significance of key mutations in the spike protein and the clarification of the molecular mechanisms by which non-structural proteins and accessory proteins mediate viral replication and immune antagonism (4, 111). In addition, in the face of rapidly evolving strains such as Omicron, ISA can efficiently conduct the analysis of variant biological characteristics, evaluation of vaccine efficacy, and assessment of antiviral drugs (8). Moreover, it has been successfully extended to various types of coronaviruses, making it a versatile and promising universal reverse genetics research tool (48).

4.5.2 BAC system

BAC vectors can stably clone large viral genomes and have been successfully used in reverse genetics studies of coronaviruses such as SARS-CoV-2 for the genetic engineering of infectious recombinant viruses (7). The introduction of key mutations including N501Y and Y453F within the RBD of the spike (S) protein allows researchers to bypass biosafety levels-3 (BSL-3) laboratory restrictions and rapidly evaluate the transmissibility and immune evasion characteristics of variant strains under lower biosafety conditions (23). Furthermore, the BAC system facilitates the rational design of live-attenuated vaccine candidates. Deletion of virulence-associated genes such as NSP1 and ORF6/7/8 contributes to the construction of viral backbones with high safety and favorable immunogenicity (13, 111). Notably, this platform enables rapid response to emerging viral variants, efficiently incorporating their genomic features into established vaccine backbones and greatly accelerating the development and translational application of next-generation broad-spectrum vaccines.

In summary, viral reverse genetics technology has evolved from a single infectious clone method to diverse optimization strategies, including ISA, BAC, and dual-promoter systems. These advancements have significantly improved the efficiency and flexibility of viral genome manipulation, laying a solid foundation for more in-depth virological research.

4.5.3 Yeast homologous recombination and TAR cloning technology

The yeast Homologous Recombination/TAR platform, based on Saccharomyces cerevisiae’s homologous recombination mechanism, has prominent advantages over traditional Escherichia coli-based cloning methods, being particularly suitable for constructing large and complex genomes like SARS-CoV-2.

In 2020, Thao et al. first reported the reconstruction of replication-competent SARS-CoV-2 using this yeast platform in Nature (10). Starting from chemically synthesized oligonucleotides, the team assembled full-length viral cDNA clones and rescued infectious live viruses within a week, significantly accelerating early functional research and vaccine development against COVID-19 (10). Subsequently, Singh et al. used multi-copy yeast vectors to assemble the SARS-CoV-2 genome, which improved the extraction efficiency of full-length viral genomes and laid a solid foundation for large-scale production and translational applications (112).

In practical applications, this platform has become one of the predominant approaches for constructing SARS-CoV-2 reverse genetic systems, enabling efficient introduction of point mutations, deletions and chimeric sequences to study the effects of key spike (S) protein mutations (D614G, N501Y, L452R) on viral transmissibility, immune evasion and pathogenic mechanisms (23, 113).

Furthermore, combined with CRISPR/Cas9 gene editing technology, it allows precise secondary modification of assembled viral genomes, establishing an integrated assembly-editing-rescue workflow that enhances experimental efficiency and technical flexibility (114). The three large-genome assembly technologies overcome the cloning toxicity of coronavirus full-length cDNA and support the rapid construction of mutant strains for variant research.

4.5.4 Low-temperature conditions ensure the stability of large-capacity virus cloning structures

Temperature-dependent stability of infectious clones represents a core challenge in molecular virology and reverse genetics research, which is particularly prominent when dealing with coronaviruses such as SARS-CoV-2 that harbor large RNA genomes (approximately 30 kb). Accumulating studies have demonstrated that culturing transformed competent cells at a low temperature range of 16–25 °C is a widely validated and highly effective strategy, which can markedly improve the genetic fidelity and structural integrity of such infectious clones. For instance, researchers commonly adopt plasmid amplification at 20–25 °C during the construction of infectious clones including icSARS-CoV-2-mNG (115).

ISA, BAC vectors and yeast homologous recombination all tackle the cloning toxicity of oversized coronavirus genomes. After cross-performance comparison, we judge that yeast recombination enables the fastest de novo assembly for rapidly mutating variant strains; BAC plasmids exhibit the highest long-term genetic stability for serial passaged recombinant viruses; ISA fragments bypass full-length plasmid construction yet yield lower viral titers and suffer from fragment cytotoxicity during amplification. Each technology requires differentiated deployment based on specific experimental objectives, and no single method dominates all research scenarios.

4.6 Exogenous gene insertion and reporter virus construction

An important application of reverse genetics is the construction of recombinant viruses expressing fluorescent proteins or reporter genes, enabling real-time monitoring of viral infection processes, replication kinetics, and the effects of viruses on cellular signaling pathways in host cells (116, 117).

4.6.1 SARS-CoV-2

Common safe insertion sites in the SARS-CoV-2 genome encompass open reading frame 7a (ORF7a), ORF8, and the 3′ untranslated region (3′UTR) downstream of the nucleocapsid (N) gene (7, 115). Deletion or modification of these regions has been proven to exert minimal impacts on viral replication in vitro, serving as ideal docking loci for the stable expression of exogenous sequences. For example, the BAC-based reverse genetic system established by Xie et al. enables efficient reporter gene insertion within the ORF7a/7b locus, achieving real-time monitoring of the viral life cycle (115). Similarly, Hou et al. utilized GFP-tagged recombinant viruses to identify distinct infection gradients of SARS-CoV-2 among different respiratory cell types, providing direct evidence for the tissue tropism of the virus (110).

4.6.2 Influenza viruses

Typical strategies for exogenous gene insertion in influenza viruses include inserting foreign genes as fusion proteins at the terminals of non-structural protein NS1, nucleoprotein (NP), or polymerase subunits PB2/PA. Alternatively, 2A self-cleaving peptides and IRES elements are widely applied to achieve polycistronic expression. A more robust approach is to construct an artificial ninth genomic segment or integrate foreign genes into the NS segment to generate bicistronic mRNA, thereby avoiding excessive disruption of viral core functions.

A large number of studies have verified the feasibility and efficacy of these engineering strategies. For instance, researchers constructed recombinant influenza viruses based on the PR8 (A/Puerto Rico/8/1934) backbone. By inserting the porcine teschovirus-1 2A (P2A) peptide and Gaussia luciferase (Gluc) coding sequence at the C-terminus of the neuraminidase (NA) gene, they obtained reporter-expressing recombinant strains PR8NAGluc/PR8 and PR8NAGluc/WSN, whose in vitro growth characteristics, genetic stability and drug susceptibility were systematically analyzed. Similarly, EGFP has been successfully integrated into the influenza virus genome and driven by the human RNA polymerase I promoter, enabling real-time visual monitoring of viral replication dynamics (118). Recombinant influenza viral vectors expressing antigens of various pathogens have also been developed, such as rFLU-P1a and rFLU-P30a carrying the dominant immunodominant regions of Mycoplasma pneumoniae P1a and P30a (119). These recombinant viruses present high genetic stability and can elicit potent specific immune responses.

Furthermore, the rgH3N2 4xM2e strain was generated by inserting multiple copies of the conserved M2 extracellular domain (M2e) epitope into the stalk region of HA. This engineered virus retains favorable growth properties and induces broad-spectrum cross-protective immunity (120). In recent studies, the SARS-CoV-2 RBD has been successfully integrated into the HA molecule or truncated NS1 protein of attenuated influenza vaccine strains, facilitating the development of bivalent candidate vaccines against both influenza and SARS-CoV-2 (83).

4.6.3 NDV

The reverse genetics system of NDV critically requires compliance with the rule of six, which stipulates that the full-length viral genome nucleotide count must be a multiple of six (19). This rule guarantees efficient recognition and normal function of the viral RNA-dependent RNA polymerase. Based on this principle, the insertion strategies for exogenous genes primarily focus on constructing additional transcription units (ATUs) within the non-coding regions of the NDV genome.

Previous studies have verified that the intergenic junction between the P and M genes (P-M junction) serves as the optimal locus for high-efficiency and stable expression of exogenous genes. For instance, recombinant NDVs have been successfully rescued by inserting the EGFP gene into the P-M junction, validating the superior performance of this insertion site (121). Moreover, the utilization of IRES downstream of the NP gene ORF represents another feasible expression strategy, offering broader options for multi-gene co-expression (122).

In vaccine development, NDV has been extensively exploited as a viral vector to express antigens of numerous clinically significant pathogens. Recombinant NDVs expressing the spike (S) protein of SARS-CoV-2 have been constructed, and animal model experiments have demonstrated that such candidates can elicit robust humoral and cellular immune responses, conferring potent protective efficacy against viral infection (123, 124). Similarly, recombinant NDVs carrying the HA gene of H9N2 avian influenza virus are capable of providing dual protection against two lethal infectious diseases in chickens (125, 126).

In the field of cancer therapy, the inherent oncolytic property of NDV, combined with exogenous gene delivery, facilitates the development of novel immune-primed oncolytic viruses. Via reverse genetic manipulation, immune-stimulatory genes such as interleukin-7 (IL-7), interleukin-15 (IL-15) and the co-stimulatory molecule OX40L have been integrated into the NDV genome, which markedly potentiates the activation of anti-tumor immune responses (127, 128). These engineered oncolytic NDVs can not only directly induce the lysis of tumor cells, but also remodel the immunosuppressive “cold” tumor microenvironment into an inflamed “hot” phenotype, thereby comprehensively improving anti-tumor therapeutic outcomes (129).

4.6.4 Report on the common inherent flaws of the virus construction system and the corresponding optimization countermeasures

Genetic instability induced by exogenous gene insertion represents a core bottleneck restricting the application of reporter viral strains and viral vectors in reverse genetics systems of RNA viruses. This phenomenon is highly conserved across all groups of RNA viruses, including positive-sense, negative-sense and double-stranded RNA viruses. Its fundamental drivers lie in the error-prone nature of viral RNA replication, the compact genomic architecture of RNA viruses, and the profound adverse impacts of inserted exogenous sequences on viral fitness. Such genetic instability manifests primarily as three major defects: recombination-mediated deletion of exogenous transgenes, loss of reporter expression resulting from cumulative point mutations, and unintended attenuation of viral virulence. These drawbacks severely compromise the reliability of reporter viruses in long-term in vivo tracing, high-throughput antiviral drug screening and novel vaccine vector development (17, 30, 31).

4.6.4.1 Fitness cost and replication defects triggered by exogenous gene insertion

Insertion of exogenous reporter genes (e.g., fluorescent protein or luciferase genes) into the RNA viral genome generally elongates the full-length genome and remodels local secondary nucleic acid structures, thereby physically impeding the replication elongation mediated by viral polymerases. When replicating extended viral genomes, RNA-dependent RNA polymerases (RdRps) frequently undergo pausing or strand slippage, which further compromises replication fidelity (130). Cumulative studies have verified that transgene insertion universally impairs viral replicative capacity, a well-documented effect termed the “fitness cost”. To compensate for this replicative disadvantage, viral populations rapidly select variant strains with exogenous fragments deleted via recombination or mutation during serial passaging, restoring their replicative competitive advantages (131, 132). For instance, in Sindbis viral vector research, unoptimized transgene expression leads to a sharp decline in viral titers. Although adaptive mutations (e.g., the G285S substitution within nsP1) acquired via directed evolution can stabilize transgene expression to a certain extent, the risk of exogenous sequence loss over long-term serial passages cannot be completely eliminated (132).

4.6.4.2 Recombination-mediated transgene deletion: the central mechanism underpinning genetic instability

RNA viral RdRp-mediated homologous recombination easily deletes inserted exogenous genes during serial passage, causing reporter loss and phenotypic drift (133). Coronaviruses, arteriviruses and picornaviruses are particularly susceptible to such recombination-induced deletion events. In porcine reproductive and respiratory syndrome virus (PRRSV), inserted exogenous genes are frequently fully eliminated by genomic recombination after continuous passaging, resulting in complete ablation of reporter signals (133, 134).

To tackle this limitation, researchers have proposed a “recombination-dependent lethal mutation” strategy: conditional lethal mutations are introduced into essential viral genes, such that only viral progeny retaining the exogenous transgene (or its flanking linker sequences) can survive, functionally enforcing the maintenance of inserted foreign sequences (135). Nevertheless, this strategy bears notable drawbacks, including drastically reduced terminal viral titers and retarded replication kinetics, which restrict its utility in large-scale virus production (136).

4.6.4.3 Non-specific interference of reporter gene expression with viral virulence and biological phenotypes

Beyond sequence-level genomic instability, constitutive expression of exogenous reporter proteins itself disrupts the native biological characteristics of viruses and triggers phenotypic drift. The folding, post-translational modification and intracellular trafficking of foreign reporter proteins consume substantial host cellular resources, and concurrently interfere with normal molecular interactions between viral structural and non-structural proteins, ultimately driving unregulated attenuation of viral virulence (137). Studies on NDV have demonstrated that transgene insertion even within non-essential genomic regions markedly attenuates viral pathogenicity in chickens. While this intrinsic attenuation benefits vaccine development, it introduces severe phenotypic bias for basic research aiming to recapitulate wild-type infection processes (137). Furthermore, fluorescent reporters (~27 kDa in molecular weight) exhibit intrinsic phototoxicity and exert steric hindrance against viral particle assembly and egress, further exacerbating genetic heterogeneity within viral populations (138, 139).

4.6.4.4 Comprehensive evaluation of current mitigation strategies and their technical limitations

Existing optimization approaches, including codon optimization, linkage of transgenes via foot-and-mouth disease virus (FMDV) 2A self-cleaving peptides, and precise genome assembly utilizing BACs, can moderately improve the initial genetic stability of reporter viruses. However, genetic drift during long-term serial passaging remains unavoidable (131, 138).

A prevailing frontier of research focuses on trans-complementation-based reporter viral systems. In this design, reporter genes are entirely separated from the viral genome and constitutively expressed by specialized helper cell lines, fundamentally eradicating genetic instability derived from genomic transgene insertion (139). Unfortunately, this system suffers from narrow applicability for in vivo animal experiments. Overall, the genetic instability inherent to reporter strains of all RNA virus families arises from an irreconcilable conflict between the intrinsic replication machinery of RNA viruses and the replicative burden imposed by exogenous inserts. Future breakthroughs in this field will likely rely on precise modulation of RdRp replication fidelity and the development of novel non-integrated expression platforms (140, 141).

Serial passaging easily triggers homologous recombination and deletion of inserted reporter genes, a universal flaw across all RNA virus reverse genetics platforms. This review proposes two innovative mitigation approaches: first, trans-complementation reporter systems utilizing stable helper cell lines to separate reporter expression from viral genomes and eliminate recombinant deletion entirely; second, recombination-dependent lethal mutations integrated into viral backbones to impose selective pressure for retaining foreign inserts. Conventional codon optimization and 2A peptide linkage can only alleviate transgene loss temporarily and fail to guarantee long-term genetic stability during continuous in vivo passaging.

5 Advances in applied research

5.1 Vaccine development

Viral reverse genetics has become the dominant platform for rational, rapid, and safe vaccine development against RNA viruses, enabling precise attenuation, antigen optimization, vector design, and rapid response to emerging variants. Unlike traditional empirical approaches, reverse genetics allows targeted genome editing, which significantly accelerates vaccine design, preclinical validation, and clinical translation (4).

5.1.1 Influenza vaccine

Reverse genetics revolutionized influenza vaccine manufacturing and pandemic preparedness by solving three historic bottlenecks: slow strain adaptation (4–6 weeks in eggs), low viral yield, and safety risks from wild-type virus handling.

The eight-plasmid reverse genetics system enables complete separation of antigen genes (HA/NA) from internal backbone genes (PB2, PB1, PA, NP, M, NS) (9). By inserting HA/NA from circulating epidemic strains into the high-growth PR8 backbone, high-yield, attenuated, antigen-matched vaccine seeds can be generated in only 5–7 days, drastically shortening the seasonal update cycle (142). This platform eliminates reliance on egg adaptation, improves batch consistency, and reduces safety hazards associated with wild-type viruses (143).

For live attenuated influenza vaccines (LAIVs), reverse genetics enables precise introduction of cold-adapted, temperature-sensitive mutations in PB2 and NS1, restricting replication to the upper respiratory tract and eliminating systemic pathogenicity (9).

Representative Marketed and Clinical Examples:

• FluMist (Fluenz, AstraZeneca): A licensed live attenuated influenza vaccine (LAIV) based on cold-adapted influenza A/B backbones, constructed and rescued via reverse genetics. It has been widely used in children and adolescents globally (142).
• Seasonal inactivated influenza vaccines (Fluzone, Sanofi): The vaccine seed strains are systematically generated using reverse genetics to ensure high yield, antigenic match, and safety (144, 145).
• Candidate pandemic vaccines: Multiple H5N1, H7N9, and H3N2 vaccine candidates in Phase I/II clinical trials were rescued via reverse genetics with attenuating mutations in PB2, NS1, or HA cleavage site (146, 147).

5.1.2 SARS-CoV-2 vaccine

Reverse genetics was decisive for accelerating rational vaccine design, antigen optimization, and neutralization assessment during the COVID-19 pandemic, transforming vaccine development from empirical trial-and-error to precise engineering (148, 149).

A key contribution was functional validation of S protein prefusion stabilization via 2P proline substitutions in live recombinant viruses, directly confirming enhanced immunogenicity—this became the core design of all mRNA and viral vector vaccines (Pfizer-BioNTech, Moderna, Janssen, AstraZeneca) (13, 29). For inactivated vaccines (Sinovac, Sinopharm), reverse genetics enabled engineering of high-growth, stable seed viruses with optimized ORF3a/ORF8 regions, solving low yield and batch inconsistency, ensuring safety and antigen consistency (4, 29, 150). Reverse genetics also supported rational attenuation by precise deletion of virulence genes (ORF3a, ORF8, NSP1), yielding candidates that replicate efficiently in vitro but are highly attenuated in vivo, inducing robust immunity without disease risk (150).

Representative Marketed and Clinical Examples:

• Inactivated SARS-CoV-2 vaccines (Sinovac, Sinopharm):Reverse genetics was used to design and rescue attenuated or high-growth seed viruses for vaccine production, ensuring safety and antigen consistency (151, 152).
• mRNA vaccines (Comirnaty/Pfizer-BioNTech; Spikevax/Moderna):Reverse genetics was critical for codon optimization, S protein stabilization (2P mutation), and antigenicity evaluation in preclinical development (153, 154).
• Viral vector vaccines (Janssen/Johnson & Johnson; AstraZeneca/Vaxzevria):Reverse genetics was applied to S protein engineering, truncation, and epitope selection for vaccine design and immunogenicity testing (155, 156).
• Live attenuated vaccine candidates (Phase I trials):Several candidates with ORF3a/ORF8 deletion or NS-like attenuation were rescued via reverse genetics, showing safe and immunogenic profiles in humans (157, 158).

5.1.3 NDV vaccine

Reverse genetics transformed NDV from an unstable laboratory vector into a safe, genetically stable, scalable vaccine platform by enforcing the rule of six and enabling precise genome engineering (19).

Traditional NDV vectors suffered from genetic instability, attenuation loss, and poor foreign antigen expression. Reverse genetics solved these issues by enabling site-specific insertion of foreign antigens into stable non-coding regions (P–M intergenic junction) while maintaining genome length as multiples of six, ensuring stable replication and high-level heterologous antigen expression (19, 121). This technology enabled development of NDV as a dual-purpose vector for veterinary vaccines and human mucosal vaccines, with intrinsic oncolytic properties for cancer immunotherapy.

Representative Veterinary and Clinical Examples:

• NDV-vectored COVID-19 vaccine candidates:Multiple candidates expressing stabilized S or RBD have completed Phase I trials, demonstrating safety and robust neutralizing antibody responses (159).
• NDV-vectored avian influenza vaccines:Commercialized veterinary vaccines expressing H5/H7 HA are widely used in poultry, generated and rescued via reverse genetics (160, 161).
• NDV-vectored cancer vaccines:Several candidates expressing tumor antigens are in Phase I/II clinical trials for melanoma and lung cancer (162).

Reverse genetics underpins the construction of inactivated, mRNA, live-attenuated and viral vector vaccine candidates targeting SARS-CoV-2, influenza and NDV. After comparing translational performances of the three prototype viral vectors, we predict that NDV mucosal delivery systems possess superior potential for multi-valent pan-vaccine development; the eight-plasmid influenza platform is most mature for rapid seasonal seed strain renewal; full-length SARS-CoV-2 infectious clones carry high biosafety risks and complex manipulation procedures, making them unsuitable for mass industrial seed production. Chimeric universal viral backbones will be the core direction for next-generation broad-spectrum vaccines.

5.2 Viral virulence & virus–host interaction

Through precise manipulation of viral genomes, reverse genetics technology has become an indispensable tool in virological research, particularly playing a key role in elucidating viral virulence mechanisms, virus-host interactions, and developing novel antiviral strategies (29).

For SARS-CoV-2, reverse genetics has uncovered critical virulence factors and immune antagonists. Recombinant SARS-CoV-2 with ORF8 deletion shows attenuated pathogenicity while inducing strong immune responses (150), E protein-deficient strains fail to assemble and release mature particles (163), NSP1 site-specific mutations (K164A/H165A) impair its host mRNA translation suppression and attenuate viral replication (164), N protein mutations in RNA-binding or oligomerization domains abrogate its ability to inhibit stress granules and restrict viral replication (165), ORF6 deletion reduces viral titer by failing to inhibit interferon-stimulated genes (with less prominent effect in Calu-3 cells) (166), ORF9b deletion confirms its broad suppression of type I and III interferons by targeting multiple innate immune pathways (167); additionally, it helps unravel viral molecular mimicry (histone mimicry) (168) and virus-triggered cell death (ORF3a deletion identifying ORF3b as a pyroptosis-inducing potassium channel) (169).

For influenza virus, reverse genetics has systematically defined the roles of structural and non-structural proteins in replication, pathogenicity, and host adaptation. The NS1 protein, a multifunctional virulence factor, antagonizes interferon production, inhibits host mRNA processing, and modulates cellular signaling; reverse genetics-derived NS1 mutants confirm these functions and identify key residues for virulence (170, 171). Polymerase subunits (PB2, PB1, PA) are critical for viral replication and host range: mutations in PB2 (E627K) enhance adaptation to mammalian cells, while PA endonuclease activity is essential for cap snatching and viral transcription (172). Moreover, reverse genetics has uncovered host factors supporting influenza replication and mechanisms of immune evasion, informing antiviral and vaccine design (173).

For NDV, reverse genetics has advanced understanding of virulence, oncolytic potential, and vector–host interactions. The fusion (F) protein cleavage site is the primary virulence determinant: velogenic strains possess multi-basic cleavage sites enabling systemic spread, while lentogenic strains with monobasic sites are restricted to respiratory tissues (174). Reverse genetics-engineered F protein mutants confirm this rule and generate attenuated vaccine strains (59). The hemagglutinin-neuraminidase (HN) protein mediates receptor binding and neuraminidase activity; HN mutations alter tropism and fusogenicity (175). Additionally, NDV’s inherent oncolytic properties are dissected via reverse genetics, revealing selective replication in tumor cells and induction of immunogenic cell death (176). Viral proteins also modulate host innate immunity, with NDV inducing robust interferon responses in normal cells while evading antiviral pathways in cancer cells (177). These findings support NDV development as a safe vaccine vector and oncolytic agent.

Site-directed knockout and point mutation via reverse genetics dissect the independent virulence function of accessory and structural proteins. We emphasize that in vitro single-protein overexpression systems cannot recapitulate the complete intracellular viral replication microenvironment; only live recombinant viruses modified by reverse genetics can accurately characterize immune antagonism and pathogenic phenotypes in authentic infection cycles. Current studies predominantly investigate single viral proteins, while synergistic immune evasion effects of multi-protein combinations lack systematic reverse genetic verification.

5.3 High-throughput antiviral screening

Reverse genetics, paired with reporter-expressing recombinant viruses, has become the gold standard for high-throughput antiviral screening and targeted drug discovery, enabling rapid identification and validation of potent candidates against RNA viruses. Unlike traditional screening relying on wild-type viruses, reverse genetics-derived reporter viruses (GFP, luciferase-tagged strains) allow real-time, quantitative measurement of viral replication, significantly accelerating the discovery of both marketed drugs and clinical-stage candidates.

Current screening assays built on live recombinant reporter viruses with full infectivity can only evaluate antiviral agents targeting viral replication, assembly and pathogenic processes, and require BSL-3 or even BSL-4 containment laboratories for highly pathogenic viruses, which greatly elevates screening costs and restricts overall throughput. Complementary non-infectious surrogate platforms established via reverse genetics, including pseudoviruses, minigenomes and multiple types of subgenomic replicons, constitute independent efficacy evaluation systems free of live virus manipulation, forming a tiered drug screening workflow with clear task division (38, 45, 47).

In the primary screening stage, pseudovirus and minigenome systems are deployed under BSL-2 conditions to conduct high-throughput profiling of thousands of candidate compounds, rapidly eliminating molecules with negligible activity and prioritizing hits targeting viral entry or RNA-dependent RNA polymerases (38, 41). In the secondary validation stage, subgenomic and episomal replicons are adopted to quantify sustained viral suppression capacity of lead compounds and monitor the emergence of drug-resistant mutations during continuous viral passaging (46). In the definitive verification stage, full-length infectious reporter viruses are utilized in matching biosafety facilities to measure half-maximal inhibitory concentration (IC₅₀), cytotoxicity and comprehensive antiviral performance across the complete viral life cycle (178). Each surrogate platform bears distinct strengths and limitations and is only applicable to specific categories of antivirals. Pseudoviruses are exclusively suitable for screening entry inhibitors and neutralizing antibodies; minigenomes are specialized for probing compounds interfering with RNP assembly and polymerase activity; subgenomic replicons excel at long-term resistance evolution research. Combined application of these non-infectious tools together with full-length infectious clones establishes an integrated evaluation pipeline, overcoming the bottlenecks of single live-virus screening systems such as stringent biosafety requirements and insufficient screening throughput, and providing a comprehensive technical framework for the discovery and preclinical assessment of broad-spectrum RNA virus antivirals.

5.3.1 SARS-CoV-2 antiviral drugs & candidates

For SARS-CoV-2, reverse genetics platforms based on BAC, ISA, and CPER have been used to construct full-length infectious clones expressing eGFP, NanoLuc, or Gaussia luciferase (Gluc) in non-essential regions such as ORF7a, ORF8, and the 3′UTR downstream of the N gene (115, 149). Reporter-labeled SARS-CoV-2 allows real-time monitoring of viral replication and high-throughput screening of small-molecule inhibitors including remdesivir, molnupiravir, and nirmatrelvir (13, 35). The reporter signal is strongly correlated with viral replication, enabling reliable calculation of half-maximal inhibitory concentration (IC₅₀) and efficient identification of antiviral candidates in a high-throughput format (8, 179). The related drugs and candidates include remdesivir, nirmatrelvir, molnupiravir, S-892216 and C10 (180–184).

5.3.2 Influenza antiviral drugs & candidates

For influenza A and B viruses, reverse genetics systems have generated reporter viruses by inserting eGFP or Gluc into the NS segment, or fusing reporter genes to NA, NP, or PB2 via 2A self-cleaving peptides (118). Recombinant influenza viruses expressing Gluc or EGFP have been used in high-throughput assays to screen antiviral compounds and evaluate drug susceptibility in vitro (63). The eighth-plasmid and dual-promoter systems support stable reporter expression, allowing automated and quantitative screening without plaque assays or qPCR detection (185). The related drugs and candidates include Oseltamivir, Baloxavir marboxil, LNA-based PSL2 inhibitor (186–188).

5.3.3 NDV antiviral drugs & candidates

For NDV, reverse genetics following the “rule of six” has achieved stable expression of eGFP or luciferase by inserting ATU at the P-M gene junction (122). Reporter NDV strains have been used to screen antiviral agents and analyze oncolytic activity in tumor cells (129). The reporter system provides a convenient and efficient tool for quantitative evaluation of viral replication and antiviral efficacy in living cells (12). The related drugs and candidates include NDV-targeted oncolytic agents (189).

The tiered workflow combining pseudoviruses, replicons and full-length reporter viruses has become the standard pipeline for antiviral compound screening. Our critical comment is that fluorescent or luciferase reporter tags alternative viral replication kinetics and introduce bias in quantitative drug efficacy measurement. Marker-free endogenous reporter knock-in viral strains will evolve into the gold standard for high-throughput antiviral evaluation in subsequent research.

6 Discussion

Despite remarkable advancements in RNA virus reverse genetics over the past decades, several critical challenges remain unresolved that restrict its broader application and translational potential. For complex, highly pathogenic, or large-genome RNA viruses, the construction of efficient and stable reverse genetics systems remains technically difficult, and infectious cDNA clones frequently suffer from instability during prokaryotic propagation (190). Meanwhile, as the technical threshold of reverse genetics continues to decrease, its associated biosafety and biosecurity risks have become an increasingly prominent global concern (14). Future research should focus on the synergistic advancement of technological innovation, application expansion, and risk management, with specific directions outlined as follows:

6.1 Optimization of technical efficiency and modularity

Further integration of cutting-edge genome-editing tools such as CRISPR/Cas systems into reverse genetics will enable simplified construction of infectious clones and precise viral genome manipulation (16, 191). CRISPR/Cas systems, particularly Cas9 and Cas13, will significantly advance RNA virus reverse genetics and infectious clone design by enabling rapid, precise genome engineering. Cas9 mediates site-specific cleavage of viral cDNA, supporting seamless insertion/deletion/substitution without restriction site dependence, which drastically accelerates assembly of large, unstable genomes like SARS-CoV-2 (~30 kb) (10, 114, 192). Cas13, an RNA-targeting CRISPR effector, directly edits viral RNA transcripts in cells, enabling transient gene knockdown or precise modification of RNA virus genomes during rescue, avoiding cDNA toxicity and improving rescue efficiency (16, 191). Together, Cas9 and Cas13 shorten cloning cycles from weeks to days and enhance fidelity for infectious clone construction.

Cas9 enables scarless introduction of key mutations (SARS-CoV-2 S-2P, influenza PB2-E627K) and high-throughput mutant library screening for virulence determinants. Cas13 complements this by targeting viral RNA to knockdown essential genes or correct defective transcripts during reverse genetics rescue, supporting rescue of difficult-to-clone viruses with toxic cDNA. Both systems streamline reporter gene insertion at safe loci (ORF7a, NS, P–M junction) and ensure recombinant viruses carry only intended edits, accelerating vaccine design and antiviral target validation (193, 194).

CRISPR-Cas9 mediates scarless viral cDNA editing, while Cas13 achieves intracellular RNA modification during virus rescue. Based on the full collation of cloning bottlenecks, we predict a short-term innovation roadmap within 5 years: integrating CRISPR scarless cleavage into yeast/BAC assembly pipelines to complete targeted mutant construction within one week. In the medium-to-long term, high-throughput viral mutant libraries edited by CRISPR will enable large-scale screening of virulence and drug-resistance loci, breaking the restriction of single-gene modification in traditional reverse genetics.

Looking ahead, standardized all-in-one CRISPR-editing assembly kits combining yeast/BAC vectors will be commercialized within 5 years, which can further cut down manual cloning operations and support parallel construction of multiple mutant virus libraries for large-scale virulence screening. Such CRISPR-integrated assembly pipelines lay a foundational technical framework for the standardized universal kits elaborated in the future prospects section.

6.2 Expanding application coverage for emerging and re-emerging viruses

Reverse genetics should be rapidly extended to understudied emerging and re-emerging viruses lacking effective prevention and control measures, such as Nipah virus (25), Ebola virus (26), Senecavirus A (195), Transmissible Gastroenteritis Virus (196) and Getah virus (197). Such expansion will accelerate the elucidation of viral biology and pathogenic mechanisms, supporting rapid development of vaccines, drugs, and intervention strategies (15). The COVID-19 pandemic has clearly demonstrated that rapid establishment of reverse genetics systems is essential for responding to sudden viral outbreaks (198).

Among these threatening pathogens, Nipah virus and Ebola virus stand out as representative high-risk zoonotic non-segmented negative-sense RNA viruses with ultra-high mortality and frequent cross-species spillover threats. Their reverse genetic platforms share the core technical framework of NDV, the canonical non-segmented -ssRNA virus analyzed throughout this review, making them ideal cases for in-depth discussion.

Reverse genetics urgently needs wider application in poorly characterized dangerous viruses including Nipah virus and Ebola virus, two high-mortality zoonotic non-segmented negative-sense RNA pathogens prone to cross-species transmission, whose reverse genetic systems share the core technical framework of NDV. Researchers have constructed Pol II-driven full-length infectious clones for various Nipah virus strains and full-length rescue platforms for Ebola virus (199–201).; multiple minigenome replicon and non-infectious life cycle simulation systems compatible with BSL-2 labs have also been developed for both viruses to bypass strict BSL-4 limits, supporting high-throughput antiviral screening, protein function research and rapid reconstruction of strains from clinical sequencing data (202–204). At the molecular level, NiV and EBOV share consistent reverse genetic workflows relying on co-transfection of full-genome plasmids and helper plasmids encoding viral structural and replicative proteins, and site-directed mutant viruses rescued by this technology greatly facilitate vaccine and drug development against the two pathogens.

Though sharing core plasmids with NDV, Nipah/Ebola genomes carry stronger prokaryotic cytotoxicity; existing low-copy vector optimization schemes for NDV cannot be directly transferred to these highly virulent pathogens. Current BSL-2 surrogate tools only reflect partial replication steps, failing to simulate cross-species transmission and systemic pathogenicity, which creates an unbridgeable gap between preliminary screening and in vivo evaluation.

Collectively, reverse genetics delivers transformative research capacity for Nipah and Ebola viruses and accelerates the development of corresponding prophylactic and therapeutic interventions. Owing to their inherent high pathogenicity, all manipulations involving intact live viruses or GOF modifications must abide by standardized biosafety protocols with continuous risk assessment and mitigation³. Establishing universal, low-containment reverse genetic platforms for Nipah and Ebola viruses to advance broad-spectrum antiviral discovery and pre-pandemic vaccine stockpiling remains a core priority for future virological research.

This review puts forward a targeted development strategy: research priorities should be placed on optimizing minigenome and non-infectious replicon surrogate systems compatible with BSL-2 laboratories to reduce reliance on high-containment facilities. Meanwhile, standardized prefabricated full-length antigenomic backbones ought to be established for these high-mortality zoonotic viruses to rapidly generate vaccine candidates upon spillover outbreaks. Targeted low-BSL surrogate platforms for BSL-4 pathogens will be prioritized core development directions outlined in Section 6.7.

6.3 Strengthened biosafety and biosecurity governance

With wider accessibility of reverse genetics, stricter biosafety protocols and operational guidelines are urgently needed to ensure responsible use and mitigate biological risks (14). For high-risk pathogens requiring BSL-4 containment (Nipah virus and Ebola virus), reverse genetics already enables safe genome manipulation without direct handling of wild-type viruses, substantially reducing laboratory hazards (25). Future efforts should focus on engineering self-attenuating, replication-defective viral systems to further enhance intrinsic safety.

As reverse genetic manipulation thresholds keep declining, the risk of laboratory leakage and unauthorized gain-of-function modification of highly pathogenic viruses rises persistently. This review raises a core original perspective: administrative laboratory grading supervision alone cannot fully eliminate biosecurity loopholes; mandatory intrinsic safety engineering elements must be embedded into all universal reverse genetic vectors, such as conditional lethal switches that terminate viral replication outside designated cell lines. Distinct ethical and risk assessment standards should be implemented to separate routine basic virology research from GOF manipulation experiments.

6.4 Standard biosafety guidelines for reconstructing RNA viruses

Recombinant RNA viruses, including SARS-CoV-2, influenza virus and NDV, are characterized by high pathogenicity, transmissibility and genomic instability. All related experimental operations must strictly comply with international and national biosafety regulations. Hierarchical biosafety management is implemented based on risk assessment to ensure that laboratory operations, personal protection and environmental control are fully adapted to the hazard level of specific pathogens.

Distinct pathogens are classified into different BSL. SARS-CoV-2 is designated as a BSL-3 agent by the WHO, CDC, and national regulations on pathogenic microorganism laboratory biosafety in China (205). It can spread efficiently via aerosols and may cause severe or fatal respiratory diseases (206). The biosafety classification of influenza viruses depends on strain virulence: seasonal influenza viruses are routinely handled under BSL-2 conditions, whereas highly pathogenic avian influenza strains such as H5N1 require enhanced BSL-3 facilities3 (207). Similarly, low-pathogenic NDV strains are manipulated at BSL-2, while highly pathogenic strains, as well as recombinant NDV (rNDV) with undefined virulence or for large-scale production, necessitate upgraded BSL-3 containment (208).

Live virus rescue from cDNA clones via reverse genetics represents the highest-risk procedure in recombinant RNA virus research. Such experiments require not only facility containment at the corresponding BSL but also rigorous ethical and biosafety review. Specifically, gain-of-function studies that may enhance viral transmissibility, immune evasion or pathogenicity must undergo dedicated risk–benefit assessment by national ethical committees, so as to prevent public health threats caused by the leakage of genetically modified pathogens. Unified international biosafety classification standards need further revision to match the rapid iteration of synthetic viral technology, a key future governance demand.

6.5 Integration with AI and big data analytics

Deep interdisciplinary fusion of reverse genetics with artificial intelligence and big data will revolutionize viral mutation prediction, vaccine efficacy evaluation, and high-throughput antiviral target discovery (209). Combined with mutant libraries constructed via reverse genetics, big-data mining can rapidly identify key loci governing viral replication, pathogenicity, and drug resistance, laying a foundation for precision prevention and control.

Existing AI tools merely predict spike mutations detached from rescue pipelines. We propose a closed-loop framework: AI predicts virulence/escape mutations → modular plasmid construction → high-throughput phenotyping → model iterative retraining to replace blind mutagenesis. The closed-loop AI-virus editing workflow will be further refined and industrialized as a long-term developmental priority in the following prospect subsection.

6.6 Accelerated development of novel vaccines and therapeutics

Reverse genetics will remain the cornerstone for developing next-generation, broadly protective RNA virus vaccines and precision antivirals, driving a paradigm shift from reactive response to proactive prevention. By enabling rational attenuation, antigen engineering, and rapid strain adaptation, this technology will accelerate the design of universal influenza vaccines targeting conserved HA stalk or M2e epitopes, as well as pan-coronavirus vaccines based on conserved S protein domains or shared epitopes (143). For emerging pathogens, reverse genetics will support one-health vaccine platforms that rapidly generate candidate vaccines within weeks of viral discovery.

In antiviral development, reverse genetics will enable targeted, mechanism-driven drug design and personalized therapeutic strategies. Combined with CRISPR/Cas and AI-driven mutagenesis, it will accelerate the identification of conserved viral targets and the development of broad-spectrum antivirals active against multiple RNA viruses. Future prospects also include live-attenuated vaccines with built-in safety switches and engineered oncolytic viruses optimized for cancer immunotherapy, positioning reverse genetics as a critical engine for global public health preparedness and innovative therapeutic development.

Reverse genetics acts as the foundational engineering tool for next-generation broad-spectrum vaccines and precision antivirals. From the perspective of translational industrialization and pandemic preparedness, we anticipate that prefabricated universal viral backbone libraries will become standard national public health reserves; NDV and influenza chimeric multi-valent vaccine vectors will advance toward commercial production; low-cost replicon-based in vitro screening platforms will drastically cut early-stage R&D expenditure for broad-spectrum antiviral compounds. Universal multi-valent viral backbones and prefabricated plasmid stockpiles represent core translational directions for the next decade, detailed below.

6.7 Future development directions

6.7.1 Short-term priorities

Future commercial modular plasmids will embed conditional lethal cassettes as standard built-in elements, unifying ribozyme-promoter assemblies and interchangeable helper fragments for +ssRNA, segmented and non-segmented −ssRNA viruses.

Marker-free small peptide knock-in technology will replace bulky fluorescent reporters to eliminate fitness interference and phenotypic bias in antiviral high-throughput assays.

CRISPR-yeast one-pot assembly modules will be packaged into unified kits to shorten mutant library construction cycles without redundant bacterial amplification steps.

6.7.2 Medium-and long-term breakthrough directions

6.7.2.1 Closed-loop artificial intelligence rational virus design platform

The future system will couple automatic transfection and detection robots to build unattended mutant screening pipelines. Beyond single spike mutation prediction, AI models will simultaneously forecast RdRp mutations and combined drug-resistance phenotypes to cut blind strain construction.

6.7.2.2 Intrinsic biosafety vector engineering

Single lethal switches carry escape risks. Dual independent conditional lethal circuits will be integrated into universal backbones for all GOF modified viruses to block accidental horizontal transmission outside designated cell lines.

6.7.2.3 Optimized low-containment systems for BSL-4 pathogens

Stable cell lines stably expressing full NP/P/L helper proteins will be engineered to eliminate repeated co-transfection, enabling continuous long-term drug evaluation under BSL-2 conditions for filoviruses and henipaviruses.

6.7.3 Translational and public health-oriented future layouts

National standardized attenuated viral backbones (influenza, NDV, coronavirus, Nipah, Ebola) will be preserved as public health reserves. Upon new zoonotic pathogen sequencing, surface antigen cassettes can be rapidly inserted to generate candidate vaccine strains within one week.

Industrialized replicon cell lines will be scaled down to lower early-stage anti-viral R&D costs, focusing on pan-viral polymerase and protease inhibitor discovery. Chimeric NDV-influenza shuttle backbones will be advanced for multi-valent single-dose vaccines against multiple zoonotic pathogens.

6.7.4 Global governance and ethical standardization perspectives

Regional inconsistent GOF review standards need unified hierarchical evaluation frameworks: routine viral functional research adopts simplified approval, while transmissibility-enhancing modification requires national-level independent risk assessment, complete experimental data archiving and regular follow-up audits.

Cross-border regulation of synthetic gene orders will be established to restrict commercial full-length cDNA synthesis of highly lethal viruses, with joint international filing for recombinant pathogen strain transportation.

6.7.5 Three-stage evolution trajectory of RNA virus reverse genetics and cross-disciplinary prospects

Most existing related reviews merely list isolated technical improvements without classifying staged developmental goals or linking technical progress to public health and industrial demands. From our systematic cross-comparison of three prototype viral reverse genetic platforms, we propose that the future evolution of this technology follows a three-stage trajectory:

Stage 1: Standardization and simplification of single-virus editing tools (short-term);

Stage 2: Interdisciplinary fusion of synthetic biology and artificial intelligence for rational virus design (medium-term);

Stage 3: Deep integration with pandemic prevention systems and global biosafety governance (long-term).

All future technical iterations must balance research efficiency and inherent biological risk control; simply pursuing higher rescue efficiency without embedded safety elements will restrict the wide clinical and public health application of reverse genetics technology.

7 Conclusion

This review systematically collates the technical evolution, core modular components, multi-scenario translational applications and unresolved bottlenecks of RNA virus reverse genetics, with three innovative integrated viewpoints proposed through cross-comparison of SARS-CoV-2, NDV and influenza A virus systems. First, a three-tier classification system of technical bottlenecks is summarized, demonstrating that single modification of promoters or vectors cannot solve all cloning defects, requiring synchronous optimization of modular plasmids, low-biosafety surrogate tools and intrinsic safety cassettes. Second, a standardized three-stage antiviral screening pipeline (pseudovirus primary screening, replicon secondary validation, full-length infectious clone final verification) is constructed to balance high experimental throughput and laboratory biosafety constraints. Third, this work predicts the long-term developmental trajectory of this field: RNA virus reverse genetics will transform from an independent single-discipline virology technology into an integrated cross-disciplinary platform combining synthetic biology, artificial intelligence, big data pandemic early warning and biosafety engineering. Future core research priorities focus on six actionable directions: the development of cross-species universal modular plasmid toolkits, CRISPR-Cas integrated rapid viral assembly pipelines, low-BSL surrogate systems for BSL-4 high-risk viruses, AI closed-loop intelligent viral design platforms, built-in intrinsic biosafety vector backbones, and industrialized universal vaccine and broad-spectrum antiviral development chains. All future technical innovations must be matched with embedded biosecurity control elements to achieve a sustainable balance between efficient viral genetic engineering and global public health risk prevention. This systematic summary provides comprehensive theoretical and technical support for responding to emerging and re-emerging RNA viral pandemics worldwide, and promotes the iterative upgrading and large-scale translational application of reverse genetics technology.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Graciela Andrei, KU Leuven, Belgium

Reviewed by: Sreekanth Gopinathan Pillai, Manipal Institute of Virology, India

Jingjing Yan, Shanghai Public Health Clinical Center affiliated to Fudan University, China