Empowering SARS-CoV-2 variant neutralization with a bifunctional antibody engineered with tandem heptad repeat 2 peptides.

With the global pandemic and the continuous mutations of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the need for effective and broadly neutralizing treatments has become increasingly urgent. This study introduces a novel strategy that targets two aspects simultaneously, using bifunctional antibodies to inhibit both the attachment of SARS-CoV-2 to host cell membranes and viral fusion. We developed pioneering IgG4-(HR2)4 bifunctional antibodies by creating immunoglobulin G4-based and phage display-derived human monoclonal antibodies (mAbs) that specifically bind to the SARS-CoV-2 receptor-binding domain, engineered with four heptad repeat 2 (HR2) peptides. Our in vitro experiments demonstrate the superior neutralization efficacy of these engineered antibodies against various SARS-CoV-2 variants, ranging from original SARS-CoV-2 strain to the recently emerged Omicron variants, as well as SARS-CoV, outperforming the parental mAb. Notably, intravenous monotherapy with the bifunctional antibody neutralizes a SARS-CoV-2 variant in a murine model without causing significant toxicity. In summary, this study unveils the significant potential of HR2 peptide-driven bifunctional antibodies as a potent and versatile strategy for mitigating SARS-CoV-2 infections. This approach offers a promising avenue for rapid development and management in the face of the continuously evolving SARS-CoV-2 variants, holding substantial promise for pandemic control.

A longitudinal molecular and cellular lung atlas of lethal SARS-CoV-2 infection in K18-hACE2 transgenic mice.

BACKGROUND: The global pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to approximately 500 million cases and 6 million deaths worldwide. Previous investigations into the pathophysiology of SARS-CoV-2 primarily focused on peripheral blood mononuclear cells from patients, lacking detailed mechanistic insights into the virus's impact on inflamed tissue. Existing animal models, such as hamster and ferret, do not faithfully replicate the severe SARS-CoV-2 infection seen in patients, underscoring the need for more relevant animal system-based research. METHODS: In this study, we employed single-cell RNA sequencing (scRNA-seq) with lung tissues from K18-hACE2 transgenic (TG) mice during SARS-CoV-2 infection. This approach allowed for a comprehensive examination of the molecular and cellular responses to the virus in lung tissue. FINDINGS: Upon SARS-CoV-2 infection, K18-hACE2 TG mice exhibited severe lung pathologies, including acute pneumonia, alveolar collapse, and immune cell infiltration. Through scRNA-seq, we identified 36 different types of cells dynamically orchestrating SARS-CoV-2-induced pathologies. Notably, SPP1+ macrophages in the myeloid compartment emerged as key drivers of severe lung inflammation and fibrosis in K18-hACE2 TG mice. Dynamic receptor-ligand interactions, involving various cell types such as immunological and bronchial cells, defined an enhanced TGFß signaling pathway linked to delayed tissue regeneration, severe lung injury, and fibrotic processes. INTERPRETATION: Our study provides a comprehensive understanding of SARS-CoV-2 pathogenesis in lung tissue, surpassing previous limitations in investigating inflamed tissues. The identified SPP1+ macrophages and the dysregulated TGFß signaling pathway offer potential targets for therapeutic intervention. Insights from this research may contribute to the development of innovative diagnostics and therapies for COVID-19. FUNDING: This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (2020M3A9I2109027, 2021R1A2C2004501).

Development of a neutralization monoclonal antibody with a broad neutralizing effect against SARS-CoV-2 variants.

BACKGROUND: The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants has challenged the effectiveness of current therapeutic regimens. Here, we aimed to develop a potent SARS-CoV-2 antibody with broad neutralizing effect by screening a scFv library with the spike protein receptor-binding domain (RBD) via phage display. METHODS: SKAI-DS84 was identified through phage display, and we performed pseudovirus neutralization assays, authentic virus neutralization assays, and in vivo neutralization efficacy evaluations. Furthermore, surface plasmon resonance (SPR) analysis was conducted to assess the physical characteristics of the antibody, including binding kinetics and measure its affinity for variant RBDs. RESULTS: The selected clones were converted to human IgG, and among them, SKAI-DS84 was selected for further analyses based on its binding affinity with the variant RBDs. Using pseudoviruses, we confirmed that SKAI-DS84 was strongly neutralizing against wild-type, B.1.617.2, B.1.1.529, and subvariants of SARS-CoV-2. We also tested the neutralizing effect of SKAI-DS84 on authentic viruses, in vivo and observed a reduction in viral replication and improved lung pathology. We performed binding and epitope mapping experiments to understand the mechanisms underlying neutralization and identified quaternary epitopes formed by the interaction between RBDs as the target of SKAI-DS84. CONCLUSIONS: We identified, produced, and tested the neutralizing effect of SKAI-DS84 antibody. Our results highlight that SKAI-DS84 could be a potential neutralizing antibody against SARS-CoV-2 and its variants.

SARS-CoV-2 infection engenders heterogeneous ribonucleoprotein interactions to impede translation elongation in the lungs.

Translational regulation in tissue environments during in vivo viral pathogenesis has rarely been studied due to the lack of translatomes from virus-infected tissues, although a series of translatome studies using in vitro cultured cells with viral infection have been reported. In this study, we exploited tissue-optimized ribosome profiling (Ribo-seq) and severe-COVID-19 model mice to establish the first temporal translation profiles of virus and host genes in the lungs during SARS-CoV-2 pathogenesis. Our datasets revealed not only previously unknown targets of translation regulation in infected tissues but also hitherto unreported molecular signatures that contribute to tissue pathology after SARS-CoV-2 infection. Specifically, we observed gradual increases in pseudoribosomal ribonucleoprotein (RNP) interactions that partially overlapped the trails of ribosomes, being likely involved in impeding translation elongation. Contemporaneously developed ribosome heterogeneity with predominantly dysregulated 5 S rRNP association supported the malfunction of elongating ribosomes. Analyses of canonical Ribo-seq reads (ribosome footprints) highlighted two obstructive characteristics to host gene expression: ribosome stalling on codons within transmembrane domain-coding regions and compromised translation of immunity- and metabolism-related genes with upregulated transcription. Our findings collectively demonstrate that the abrogation of translation integrity may be one of the most critical factors contributing to pathogenesis after SARS-CoV-2 infection of tissues.

A novel bispecific antibody dual-targeting approach for enhanced neutralization against fast-evolving SARS-CoV-2 variants.

Introduction: The emergence of new severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants has caused unprecedented health and socioeconomic crises, necessitating the immediate development of highly effective neutralizing antibodies. Despite recent advancements in anti-SARS-CoV-2 receptor-binding domain (RBD)-specific monoclonal antibodies (mAbs) derived from convalescent patient samples, their efficacy against emerging variants has been limited. In this study, we present a novel dual-targeting strategy using bispecific antibodies (bsAbs) that specifically recognize both the SARS-CoV-2 RBD and fusion peptide (FP), crucial domains for viral attachment to the host cell membrane and fusion in SARS-CoV-2 infection. Methods: Using phage display technology, we rapidly isolated FP-specific mAbs from an established human recombinant antibody library, identifying K107.1 with a nanomolar affinity for SARS-CoV-2 FP. Furthermore, we generated K203.A, a new bsAb built in immunoglobulin G4-(single-chain variable fragment)2 forms and demonstrating a high manufacturing yield and nanomolar affinity to both the RBD and FP, by fusing K102.1, our previously reported RBD-specific mAb, with K107.1. Results: Our comprehensive in vitro functional analyses revealed that the K203.A bsAb significantly outperformed the parental RBD-specific mAb in terms of neutralization efficacy against SARS-CoV-2 variants. Furthermore, intravenous monotherapy with K203.A demonstrated potent in vivo neutralizing activity without significant in vivo toxicity in a mouse model infected with a SARS-CoV-2 variant. Conclusion: These findings present a novel bsAb dual-targeting strategy, directed at SARS-CoV-2 RBD and FP, as an effective approach for rapid development and management against continuously evolving SARS-CoV-2 variants.

Novel bispecific human antibody platform specifically targeting a fully open spike conformation potently neutralizes multiple SARS-CoV-2 variants.

Rapid emergence of new variants of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has prompted an urgent need for the development of broadly applicable and potently neutralizing antibody platform against the SARS-CoV-2, which can be used for combatting the coronavirus disease 2019 (COVID-19). In this study, based on a noncompeting pair of phage display-derived human monoclonal antibodies (mAbs) specific to the receptor-binding domain (RBD) of SARS-CoV-2 isolated from human synthetic antibody library, we generated K202.B, a novel engineered bispecific antibody with an immunoglobulin G4-single-chain variable fragment design, with sub- or low nanomolar antigen-binding avidity. Compared with the parental mAbs or mAb cocktail, the K202.B antibody showed superior neutralizing potential against a variety of SARS-CoV-2 variants in vitro. Furthermore, structural analysis of bispecific antibody-antigen complexes using cryo-electron microscopy revealed the mode of action of K202.B complexed with a fully open three-RBD-up conformation of SARS-CoV-2 trimeric spike proteins by simultaneously interconnecting two independent epitopes of the SARS-CoV-2 RBD via inter-protomer interactions. Intravenous monotherapy using K202.B exhibited potent neutralizing activity in SARS-CoV-2 wild-type- and B.1.617.2 variant-infected mouse models, without significant toxicity in vivo. The results indicate that this novel approach of development of immunoglobulin G4-based bispecific antibody from an established human recombinant antibody library is likely to be an effective strategy for the rapid development of bispecific antibodies, and timely management against fast-evolving SARS-CoV-2 variants.

Establishment of multicenter COVID-19 therapeutics preclinical test system in Republic of Korea.

Throughout the recent COVID-19 pandemic, South Korea led national efforts to develop vaccines and therapeutics for SARS-CoV-2. The project proceeded as follows: 1) evaluation system setup (including Animal Biosafety Level 3 (ABSL3) facility alliance, standardized nonclinical evaluation protocol, and laboratory information management system), 2) application (including committee review and selection), and 3) evaluation (including expert judgment and reporting). After receiving 101 applications, the selection committee reviewed pharmacokinetics, toxicity, and efficacy data and selected 32 final candidates. In the nonclinical efficacy test, we used golden Syrian hamsters and human angiotensin-converting enzyme 2 transgenic mice under a cytokeratin 18 promoter to evaluate mortality, clinical signs, body weight, viral titer, neutralizing antibody presence, and histopathology. These data indicated eight new drugs and one repositioned drug having significant efficacy for COVID-19. Three vaccine and four antiviral drugs exerted significant protective activities against SARS-CoV-2 pathogenesis. Additionally, two anti-inflammatory drugs showed therapeutic effects on lung lesions and weight loss through their mechanism of action but did not affect viral replication. Along with systematic verification of COVID-19 animal models through large-scale studies, our findings suggest that ABSL3 multicenter alliance and nonclinical evaluation protocol standardization can promote reliable efficacy testing against COVID-19, thus expediting medical product development.

Corrigendum: Mouse models of lung-specific SARS-CoV-2 infection with moderate pathological traits.

[This corrects the article DOI: 10.3389/fimmu.2022.1055811.].

Mouse models of lung-specific SARS-CoV-2 infection with moderate pathological traits.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causing coronavirus disease 2019 (COVID-19) has been a global health concern since 2019. The viral spike protein infects the host by binding to angiotensin-converting enzyme 2 (ACE2) expressed on the cell surface, which is then processed by type II transmembrane serine protease. However, ACE2 does not react to SARS-CoV-2 in inbred wild-type mice, which poses a challenge for preclinical research with animal models, necessitating a human ACE2 (hACE2)-expressing transgenic mouse model. Cytokeratin 18 (K18) promoter-derived hACE2 transgenic mice [B6.Cg-Tg(K18-ACE2)2Prlmn/J] are widely used for research on SARS-CoV-1, MERS-CoV, and SARS-CoV-2. However, SARS-CoV-2 infection is lethal at ≥105 PFU and SARS-CoV-2 target cells are limited to type-1 alveolar pneumocytes in K18-hACE2 mice, making this model incompatible with infections in the human lung. Hence, we developed lung-specific SARS-CoV-2 infection mouse models with surfactant protein B (SFTPB) and secretoglobin family 1a member 1 (Scgb1a1) promoters. After inoculation of 105 PFU of SARS-CoV-2 to the K18-hACE2, SFTPB-hACE2, and SCGB1A1-hACE2 models, the peak viral titer was detected at 2 days post-infection and then gradually decreased. In K18-hACE2 mice, the body temperature decreased by approximately 10°C, body weight decreased by over 20%, and the survival rate was reduced. However, SFTPB-hACE2 and SCGB1A1-hACE2 mice showed minimal clinical signs after infection. The virus targeted type I pneumocytes in K18-hACE2 mice; type II pneumocytes in SFTPB-hACE2 mice; and club, goblet, and ciliated cells in SCGB1A1-hACE2 mice. A time-dependent increase in severe lung lesions was detected in K18-hACE2 mice, whereas mild lesions developed in SFTPB-hACE2 and SCGB1A1-hACE2 mice. Spleen, small intestine, and brain lesions developed in K18-hACE2 mice but not in SFTPB-hACE2 and SCGB1A1-hACE2 mice. These newly developed SFTPB-hACE2 and SCGB1A1-hACE2 mice should prove useful to expand research on hACE2-mediated respiratory viruses.

Temporal Transcriptome Analysis of SARS-CoV-2-Infected Lung and Spleen in Human ACE2-Transgenic Mice.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a highly transmissible and potentially fatal virus. So far, most comprehensive analyses encompassing clinical and transcriptional manifestation have concentrated on the lungs. Here, we confirmed evident signs of viral infection in the lungs and spleen of SARS-CoV-2-infected K18-hACE2 mice, which replicate the phenotype and infection symptoms in hospitalized humans. Seven days post viral detection in organs, infected mice showed decreased vital signs, leading to death. Bronchopneumonia due to infiltration of leukocytes in the lungs and reduction in the spleen lymphocyte region were observed. Transcriptome profiling implicated the meticulous regulation of distress and recovery from cytokine-mediated immunity by distinct immune cell types in a time-dependent manner. In lungs, the chemokine-driven response to viral invasion was highly elevated at 2 days post infection (dpi). In late infection, diseased lungs, post the innate immune process, showed recovery signs. The spleen established an even more immediate line of defense than the lungs, and the cytokine expression profile dropped at 7 dpi. At 5 dpi, spleen samples diverged into two distinct groups with different transcriptome profile and pathophysiology. Inhibition of consecutive host cell viral entry and massive immunoglobulin production and proteolysis inhibition seemed that one group endeavored to survive, while the other group struggled with developmental regeneration against consistent viral intrusion through the replication cycle. Our results may contribute to improved understanding of the longitudinal response to viral infection and development of potential therapeutics for hospitalized patients affected by SARS-CoV-2.

Comparison of the pathogenesis of SARS-CoV-2 infection in K18-hACE2 mouse and Syrian golden hamster models.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the etiological agent of COVID-19, causes life-threatening disease. This novel coronavirus enters host cells via the respiratory tract, promoting the formation of severe pulmonary lesions and systemic disease. Few animal models can simulate the clinical signs and pathology of COVID-19 patients. Diverse preclinical studies using K18-hACE2 mice and Syrian golden hamsters, which are highly permissive to SARS-CoV-2 in the respiratory tract, are emerging; however, the systemic pathogenesis and cellular tropism of these models remain obscure. We intranasally infected K18-hACE2 mice and Syrian golden hamsters with SARS-CoV-2, and compared the clinical features, pathogenesis, cellular tropism and infiltrated immune-cell subsets. In K18-hACE2 mice, SARS-CoV-2 persistently replicated in alveolar cells and caused pulmonary and extrapulmonary disease, resulting in fatal outcomes. Conversely, in Syrian golden hamsters, transient SARS-CoV-2 infection in bronchial cells caused reversible pulmonary disease, without mortality. Our findings provide comprehensive insights into the pathogenic spectrum of COVID-19 using preclinical models.

Laboratory information management system for COVID-19 non-clinical efficacy trial data.

BACKGROUND: As the number of large-scale studies involving multiple organizations producing data has steadily increased, an integrated system for a common interoperable format is needed. In response to the coronavirus disease 2019 (COVID-19) pandemic, a number of global efforts are underway to develop vaccines and therapeutics. We are therefore observing an explosion in the proliferation of COVID-19 data, and interoperability is highly requested in multiple institutions participating simultaneously in COVID-19 pandemic research. RESULTS: In this study, a laboratory information management system (LIMS) approach has been adopted to systemically manage various COVID-19 non-clinical trial data, including mortality, clinical signs, body weight, body temperature, organ weights, viral titer (viral replication and viral RNA), and multiorgan histopathology, from multiple institutions based on a web interface. The main aim of the implemented system is to integrate, standardize, and organize data collected from laboratories in multiple institutes for COVID-19 non-clinical efficacy testings. Six animal biosafety level 3 institutions proved the feasibility of our system. Substantial benefits were shown by maximizing collaborative high-quality non-clinical research. CONCLUSIONS: This LIMS platform can be used for future outbreaks, leading to accelerated medical product development through the systematic management of extensive data from non-clinical animal studies.

Selective retinol production by modulating the composition of retinoids from metabolically engineered E. coli.

Retinoids can be produced from E. coli when introduced with the ß-carotene biosynthesis pathway and the BCMO gene. E. coli has no inherent metabolic pathways related to retinoids, therefore only retinal should be produced from the cleavage of ß-carotene by BCMO. However, retinol and retinyl acetate were also produced in significant amounts, by the non-specific activity of inherent promiscuous enzymes or the antibiotic resistance marker of the retinal-producing plasmids. Retinol was produced by the ybbO gene of E. coli which encodes oxidoreductase and retinyl acetate was produced by the chloramphenicol resistance gene, called cat gene which encodes chloramphenicol acetyltransferase, present within the pS-NA plasmid that also contains the mevalonate pathway. The composition of retinoids could be modulated by manipulating the relevant genes. The composition of retinol, a commercially important retinoid, was significantly increased by the overexpression of ybbO gene and the removal of cat gene in the recombinant E. coli, which suggests the possibility of selective retinoid production in the future.

Engineering Escherichia coli for selective geraniol production with minimized endogenous dehydrogenation.

Geraniol, a monoterpene alcohol, has versatile applications in the fragrance industry, pharmacy and agrochemistry. Moreover, geraniol could be an ideal gasoline alternative. In this study, recombinant overexpression of geranyl diphosphate synthase and the bottom portion of a foreign mevalonate pathway in Escherichia coli MG1655 produced 13.3mg/L of geraniol. Introduction of Ocimum basilicum geraniol synthase increased geraniol production to 105.2mg/L. However, geraniol production encountered a loss from its endogenous dehydrogenization and isomerization into other geranoids (nerol, neral and geranial). Three E. coli enzymes (YjgB, YahK and YddN) were identified with high sequence identity to plant geraniol dehydrogenases. YjgB was demonstrated to be the major one responsible for geraniol dehydrogenization. Deletion of yjgB increased geraniol production to 129.7mg/L. Introduction of the whole mevalonate pathway for enhanced building block synthesis from endogenously synthesized mevalonate improved geraniol production up to 182.5mg/L in the yjgB mutant after 48h of culture, which was a double of that obtained in the wild type control (96.5mg/L). Our strategy for improving geraniol production in engineered E. coli should be generalizable for addressing similar problems during metabolic engineering.

Comparison of extraction phases for a two-phase culture of a recombinant E. coli producing retinoids.

To prevent degradation of intracellular retinoids through in situ extraction from the cells, a two-phase culture system was performed. Several organic solvents, including n-alkanes, mineral oils and cosmetic raw materials, were applied as the extraction phase. Of the n-alkanes, n-decane had the highest retinoid production as 134 mg/l after 72 h. For mineral oil, light and heavy mineral oil gave retinoid productions of 158 and 174 mg/l after 96 h, respectively. Of other materials, isopropyl myristate gave the highest retinoid production of 181 mg/l. These results indicate that many types of oils can be applied for retinoid production, and optimization of the in situ extraction process will lead to further improve of economical production for the industrial purpose.

Engineered heterologous FPP synthases-mediated Z,E-FPP synthesis in E. coli.

Production of Z-type farnesyl diphosphate (FPP) has not been reported in Escherichia coli. Here we present the fusion enzyme (ILRv) of E. coli E,E-FPP synthase (IspA) and Mycobacterium tuberculosis Z,E-FPP synthase (Rv1086), which can produce primarily Z,E-FPP rather than E,E-FPP, the predominant stereoisomer found in most organisms. Z,E-farnesol (FOH) was produced from E. coli harboring the bottom portion of the MVA pathway and the fusion FPP synthase (ILRv) at a titer of 115.6 mg/L in 2YT medium containing 1% (v/v) glycerol as a carbon source and 5 mM mevalonate. The Z,E-FOH production was improved by 15-fold, compared with 7.7 mg/L obtained from the co-overexpression of separate IspA and Rv1086. The Z,E-FPP was not metabolized in native metabolic pathways of E. coli. It would be of interest to produce Z,E-FPP derived sesquiterpenes from recombinant E. coli due to no loss of Z,E-FPP substrate in endogenous metabolism of the host strain.

De novo transcriptome sequencing of Momordica cochinchinensis to identify genes involved in the carotenoid biosynthesis.

The ripe fruit of Momordica cochinchinensis Spreng, known as gac, is featured by very high carotenoid content. Although this plant might be a good resource for carotenoid metabolic engineering, so far, the genes involved in the carotenoid metabolic pathways in gac were unidentified due to lack of genomic information in the public database. In order to expedite the process of gene discovery, we have undertaken Illumina deep sequencing of mRNA prepared from aril of gac fruit. From 51,446,670 high-quality reads, we obtained 81,404 assembled unigenes with average length of 388 base pairs. At the protein level, gac aril transcripts showed about 81.5% similarity with cucumber proteomes. In addition 17,104 unigenes have been assigned to specific metabolic pathways in Kyoto Encyclopedia of Genes and Genomes, and all of known enzymes involved in terpenoid backbones biosynthetic and carotenoid biosynthetic pathways were also identified in our library. To analyze the relationship between putative carotenoid biosynthesis genes and alteration of carotenoid content during fruit ripening, digital gene expression analysis was performed on three different ripening stages of aril. This study has revealed putative phytoene synthase, 15-cis-phytone desaturase, zeta-carotene desaturase, carotenoid isomerase and lycopene epsilon cyclase might be key factors for controlling carotenoid contents during aril ripening. Taken together, this study has also made availability of a large gene database. This unique information for gac gene discovery would be helpful to facilitate functional studies for improving carotenoid quantities.

Metabolic engineering of Escherichia coli for α-farnesene production.

Sesquiterpenes are important materials in pharmaceuticals and industry. Metabolic engineering has been successfully used to produce these valuable compounds in microbial hosts. However, the microbial potential of sesquiterpene production is limited by the poor heterologous expression of plant sesquiterpene synthases and the deficient FPP precursor supply. In this study, we engineered E. coli to produce α-farnesene using a codon-optimized α-farnesene synthase and an exogenous MVA pathway. Codon optimization of α-farnesene synthase improved both the synthase expression and α-farnesene production. Augmentation of the metabolic flux for FPP synthesis conferred a 1.6- to 48.0-fold increase in α-farnesene production. An additional increase in α-farnesene production was achieved by the protein fusion of FPP synthase and α-farnesene synthase. The engineered E. coli strain was able to produce 380.0 mg/L of α-farnesene, which is an approximately 317-fold increase over the initial production of 1.2 mg/L.

Retinoid production using metabolically engineered Escherichia coli with a two-phase culture system.

BACKGROUND: Retinoids are lipophilic isoprenoids composed of a cyclic group and a linear chain with a hydrophilic end group. These compounds include retinol, retinal, retinoic acid, retinyl esters, and various derivatives of these structures. Retinoids are used as cosmetic agents and effective pharmaceuticals for skin diseases. Retinal, an immediate precursor of retinoids, is derived by ß-carotene 15,15'-mono(di)oxygenase (BCM(D)O) from ß-carotene, which is synthesized from the isoprenoid building blocks isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP). Retinoids are chemically unstable and biologically degraded via retinoic acid. Although extensive studies have been performed on the microbial production of carotenoids, retinoid production using microbial metabolic engineering has not been reported. Here, we report retinoid production using engineered Escherichia coli that express exogenous BCM(D)O and the mevalonate (MVA) pathway for the building blocks synthesis in combination with a two-phase culture system using a dodecane overlay. RESULTS: Among the BCM(D)O tested in E. coli, the synthetic retinoid synthesis protein (SR), based on bacteriorhodopsin-related protein-like homolog (Blh) of the uncultured marine bacteria 66A03, showed the highest ß-carotene cleavage activity with no residual intracellular ß-carotene. By introducing the exogenous MVA pathway, 8.7 mg/L of retinal was produced, which is 4-fold higher production than that of augmenting the MEP pathway (dxs overexpression). There was a large gap between retinal production and ß-carotene consumption using the exogenous MVA pathway; therefore, the retinal derivatives were analyzed. The derivatives, except for retinoic acid, that formed were identified, and the levels of retinal, retinol, and retinyl acetate were measured. Amounts as high as 95 mg/L retinoids were obtained from engineered E. coli DH5α harboring the synthetic SR gene and the exogenous MVA pathway in addition to dxs overexpression, which were cultured at 29°C for 72 hours with 2YT medium containing 2.0% (w/v) glycerol as the main carbon source. However, a significant level of intracellular degradation of the retinoids was also observed in the culture. To prevent degradation of the intracellular retinoids through in situ extraction from the cells, a two-phase culture system with dodecane was used. The highest level of retinoid production (136 mg/L) was obtained after 72 hours with 5 mL of dodecane overlaid on a 5 mL culture. CONCLUSIONS: In this study, we successfully produced 136 mg/L retinoids, which were composed of 67 mg/L retinal, 54 mg/L retinol, and 15 mg/L retinyl acetate, using a two-phase culture system with dodecane, which produced 68-fold more retinoids than the initial level of production (2.2 mg/L). Our results demonstrate the potential use of E. coli as a promising microbial cell factory for retinoid production.