mRNA-LNP vaccine-induced CD8+ T cells protect mice from lethal SARS-CoV-2 infection in the absence of specific antibodies

The role of CD8+ T cells in SARS-CoV-2 pathogenesis or mRNA-LNP vaccine-induced protection from lethal COVID-19 is unclear. Using mouse-adapted SARS-CoV-2 virus (MA30) in C57BL/6 mice, we show that CD8+ T cells are unnecessary for the intrinsic resistance of female or the susceptibility of male mice to lethal SARS-CoV-2 infection. Also, mice immunized with a di-proline prefusion-stabilized full-length SARS-CoV-2 Spike (S-2P) mRNA-LNP vaccine, which induces Spike-specific antibodies and CD8+ T cells specific for the Spike-derived VNFNFNGL peptide, are protected from SARS-CoV-2 infection-induced lethality and weight loss, while mice vaccinated with mRNA-LNPs encoding only VNFNFNGL are protected from lethality but not weight loss. CD8+ T cell depletion ablates protection in VNFNFNGL but not in S-2P mRNA-LNP-vaccinated mice. Therefore, mRNA-LNP vaccine-induced CD8+ T cells are dispensable when protective antibodies are present but essential for survival in their absence. Hence, vaccine-induced CD8+ T cells may be critical to protect against SARS-CoV-2 variants that mutate epitopes targeted by protective antibodies.

Multiple and Consecutive Genome Editing Using i-GONAD and Breeding Enrichment Facilitates the Production of Genetically Modified Mice

Genetically modified (GM) mice are essential tools in biomedical research. Traditional methods for generating GM mice are expensive and require specialized personnel and equipment. The use of clustered regularly interspaced short palindromic repeats (CRISPR) coupled with improved-Genome editing via Oviductal Nucleic Acids Delivery (i-GONAD) has highly increased the feasibility of producing GM mice in research laboratories. However, genetic modification in inbred mouse strains of interest such as C57BL/6 (B6) is still challenging because of their low fertility and embryo fragility. We have successfully generated multiple novel GM mouse strains in the B6 background while attempting to optimize i-GONAD. We found that i-GONAD reduced the litter size in superovulated pregnant females but did not impact pregnancy rates. Natural mating or low-hormone dose did not increase the low fertility rate observed in superovulated B6 females. However, diet enrichment had a positive effect on pregnancy success. We also optimized breeding conditions to increase the survival of small litters by co-housing i-GONAD-treated pregnant B6 females with synchronized pregnant FVB/NJ companion mothers. Thus, GM mice generation was increased by an enriched diet and shared pup rearing with highly fertile females such as FVB/NJ. In the present study, we generated 16 GM mice using a CRISPR/Cas system to target individual and multiple loci simultaneously or consecutively. We also compared homology-directed repair efficiency using different methods for LoxP insertion for conditional knockout mouse production. We found that a two-step serial LoxP insertion, in which each LoxP sequence was inserted individually in different i-GONAD procedures, was a low-risk high-efficiency method for generating floxed mice.

Design and preclinical evaluation of a universal SARS-CoV-2 mRNA vaccine

Because of the rapid mutations of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), an effective vaccine against SARS-CoV-2 variants is needed to prevent coronavirus disease 2019 (COVID-19). T cells, in addition to neutralizing antibodies, are an important component of naturally acquired protective immunity, and a number of studies have shown that T cells induced by natural infection or vaccination contribute significantly to protection against several viral infections including SARS-CoV-2. However, it has never been tested whether a T cell-inducing vaccine can provide significant protection against SARS-CoV-2 infection in the absence of preexisting antibodies. In this study, we designed and evaluated lipid nanoparticle (LNP) formulated mRNA vaccines that induce only T cell responses or both T cell and neutralizing antibody responses by using two mRNAs. One mRNA encodes SARS-CoV-2 Omicron Spike protein in prefusion conformation for induction of neutralizing antibodies. The other mRNA encodes over one hundred T cell epitopes (multi-T cell epitope or MTE) derived from non-Spike but conserved regions of the SARS-CoV-2. We show immunization with MTE mRNA alone protected mice from lethal challenge with the SARS-CoV-2 Delta variant or a mouse-adapted virus MA30. Immunization with both mRNAs induced the best protection with the lowest viral titer in the lung. These results demonstrate that induction of T cell responses, in the absence of preexisting antibodies, is sufficient to confer protection against severe disease, and that a vaccine containing mRNAs encoding both the Spike and MTE could be further developed as a universal SARS-CoV-2 vaccine.

Interferon partly dictates a divergent transcriptional response in poxvirus-infected and bystander inflammatory monocytes

Inflammatory monocytes (iMOs) and B cells are the main targets of the poxvirus ectromelia virus (ECTV) in the lymph nodes of mice and play distinct roles in surviving the infection. Infected and bystander iMOs control ECTV's systemic spread, preventing early death, while B cells make antibodies that eliminate ECTV. Our work demonstrates that within an infected animal that survives ECTV infection, intrinsic and bystander infection of iMOs and B cells differentially control the transcription of genes important for immune cell function and, perhaps, cell identity. Bystander cells upregulate metabolism, antigen presentation, and interferon-stimulated genes. Infected cells downregulate many cell-type-specific genes and upregulate transcripts typical of non-immune cells. Bystander (Bys) and infected (Inf) iMOs non-redundantly contribute to the cytokine milieu and the interferon response. Furthermore, we uncover how type I interferon (IFN-I) or IFN-γ signaling differentially regulates immune pathways in Inf and Bys iMOs and that, at steady state, IFN-I primes iMOs for rapid IFN-I production and antigen presentation.

IFN-β Acts on Monocytes to Ameliorate CNS Autoimmunity by Inhibiting Proinflammatory Cross-Talk Between Monocytes and Th Cells

IFN-β has been the treatment for multiple sclerosis (MS) for almost three decades, but understanding the mechanisms underlying its beneficial effects remains incomplete. We have shown that MS patients have increased numbers of GM-CSF+ Th cells in circulation, and that IFN-β therapy reduces their numbers. GM-CSF expression by myelin-specific Th cells is essential for the development of experimental autoimmune encephalomyelitis (EAE), an animal model of MS. These findings suggested that IFN-β therapy may function via suppression of GM-CSF production by Th cells. In the current study, we elucidated a feedback loop between monocytes and Th cells that amplifies autoimmune neuroinflammation, and found that IFN-β therapy ameliorates central nervous system (CNS) autoimmunity by inhibiting this proinflammatory loop. IFN-β suppressed GM-CSF production in Th cells indirectly by acting on monocytes, and IFN-β signaling in monocytes was required for EAE suppression. IFN-β increased IL-10 expression by monocytes, and IL-10 was required for the suppressive effects of IFN-β. IFN-β treatment suppressed IL-1β expression by monocytes in the CNS of mice with EAE. GM-CSF from Th cells induced IL-1β production by monocytes, and, in a positive feedback loop, IL-1β augmented GM-CSF production by Th cells. In addition to GM-CSF, TNF and FASL expression by Th cells was also necessary for IL-1β production by monocyte. IFN-β inhibited GM-CSF, TNF, and FASL expression by Th cells to suppress IL-1β secretion by monocytes. Overall, our study describes a positive feedback loop involving several Th cell- and monocyte-derived molecules, and IFN-β actions on monocytes disrupting this proinflammatory loop.

Resistance to lethal ectromelia virus infection requires Type I interferon receptor in natural killer cells and monocytes but not in adaptive immune or parenchymal cells

Type I interferons (IFN-I) are antiviral cytokines that signal through the ubiquitous IFN-I receptor (IFNAR). Following footpad infection with ectromelia virus (ECTV), a mouse-specific pathogen, C57BL/6 (B6) mice survive without disease, while B6 mice broadly deficient in IFNAR succumb rapidly. We now show that for survival to ECTV, only hematopoietic cells require IFNAR expression. Survival to ECTV specifically requires IFNAR in both natural killer (NK) cells and monocytes. However, intrinsic IFNAR signaling is not essential for adaptive immune cell responses or to directly protect non-hematopoietic cells such as hepatocytes, which are principal ECTV targets. Mechanistically, IFNAR-deficient NK cells have reduced cytolytic function, while lack of IFNAR in monocytes dampens IFN-I production and hastens virus dissemination. Thus, during a pathogenic viral infection, IFN-I coordinates innate immunity by stimulating monocytes in a positive feedback loop and by inducing NK cell cytolytic function.

Cutting Edge: Protection by Antiviral Memory CD8 T Cells Requires Rapidly Produced Antigen in Large Amounts

Numerous attempts to produce antiviral vaccines by harnessing memory CD8 T cells have failed. A barrier to progress is that we do not know what makes an Ag a viable target of protective CD8 T cell memory. We found that in mice susceptible to lethal mousepox (the mouse homolog of human smallpox), a dendritic cell vaccine that induced memory CD8 T cells fully protected mice when the infecting virus produced Ag in large quantities and with rapid kinetics. Protection did not occur when the Ag was produced in low amounts, even with rapid kinetics, and protection was only partial when the Ag was produced in large quantities but with slow kinetics. Hence, the amount and timing of Ag expression appear to be key determinants of memory CD8 T cell antiviral protective immunity. These findings may have important implications for vaccine design.

A pro-survival role for the intracellular granzyme B inhibitor Serpinb9 in natural killer cells during poxvirus infection

Intracellular serpins are proposed to inactivate proteases released from lysosome-related organelles into the host cell interior, preventing cell death. Serpinb9 opposes the immune cytotoxic protease, granzyme B, and in a number of settings protects cells against granzyme B-mediated cell death. Using a knockout mouse line engineered to express green fluorescent protein under the serpbinb9 promoter, we demonstrate that serpinb9 is vital for host survival during Ectromelia virus infection by maintaining both mature natural killer NK) cells, and activated CD8⁺ T cells. Serpinb9 expression parallels granzyme B expression within both populations during infection. Maturing serpinb9-null NK cells exhibit higher levels of granzyme B-mediated apoptosis during infection; hence there are fewer mature NK cells, and these cells also have lower cytotoxic potential. Thus the serpinb9-granzyme B axis is important for homeostasis of both major cytotoxic effector cell populations.

Ectromelia virus N1L is essential for virulence but not dissemination in a classical model of mousepox

Mousepox is caused by the orthopoxvirus ectromelia virus (ECTV), and is thought to be transmitted via skin abrasions. We studied the ECTV virulence factor N1 following subcutaneous infection of mousepox-susceptible BALB/c mice. In this model, ECTV lacking N1L gene was attenuated more than 1000-fold compared with wild-type virus and replication was profoundly reduced as early as four days after infection. However, in contrast to data from an intranasal model, N1 protein was not required for virus dissemination. Further, neither T cell nor cytokine responses were enhanced in the absence of N1. Together with the early timing of reduced virus titres, this suggests that in a cutaneous model, N1 exerts its function at the level of infected cells or in the inhibition of the very earliest effectors of innate immunity.

Engineering recombinant poxviruses using a compact GFP-blasticidin resistance fusion gene for selection

Recombinant poxviruses are important tools for research and some are candidate vaccines. To make these viruses a simple, small vector that can be used to engineer multiple strains of vaccinia virus and other model poxviruses, including ectromelia virus is of value. Here a set of plasmids and methods for making these viruses that uses an enhanced green fluorescent protein-blasticidin resistance (GFP-bsd) fusion gene as a transient selectable marker are described. This gene is smaller than any of the bi-functional selection markers used previously. The versatility of the method across different poxviruses is demonstrated by engineering changes into multiple loci of the WR and Modified Vaccinia Ankara (MVA) strains of vaccinia virus and also ectromelia virus. Finally, a set of vaccinia virus sequences for directing homologous recombination that are very highly conserved was designed and tested. These sequences allow a single plasmid to be used to insert a transgene into multiple strains of the virus.