There is an overall lack of knowledge about how vaccines work. Here we propose the combined use of reverse genetics and bone marrow chimeric mice to gain insight into the early host immune responses to vaccines with a special focus on dendritic cells and T cell immunity.
Vaccines are one of the greatest achievements of mankind, and have saved millions of lives over the last century. Paradoxically, little is known about the physiological mechanisms that mediate immune responses to vaccines perhaps due to the overall success of vaccination, which has reduced interest into the molecular and physiological mechanisms of vaccine immunity. However, several important human pathogens including influenza virus still pose a challenge for vaccination, and may benefit from immune-based strategies.
Although influenza reverse genetics has been successfully applied to the generation of live-attenuated influenza vaccines (LAIVs), the addition of molecular tools in vaccine preparations such as tracer components to follow up the kinetics of vaccination in vivo, has not been addressed. In addition, the recent generation of mouse models that allow specific depletion of leukocytes during kinetic studies has opened a window of opportunity to understand the basic immune mechanisms underlying vaccine-elicited protection. Here, we describe how the combination of reverse genetics and chimeric mouse models may help to provide new insights into how vaccines work at physiological and molecular levels, using as example a recombinant, cold-adapted, live-attenuated influenza vaccine (LAIV). We utilized laboratory-generated LAIVs harboring cell tracers as well as competitive bone marrow chimeras (BMCs) to determine the early kinetics of vaccine immunity and the main physiological mechanisms responsible for the initiation of vaccine-specific adaptive immunity. In addition, we show how this technique may facilitate gene function studies in single animals during immune responses to vaccines. We propose that this technique can be applied to improve current prophylactic strategies against pathogens for which urgent medical countermeasures are needed, for example influenza, HIV, Plasmodium, and hemorrhagic fever viruses such as Ebola virus.
The generation of immunological memory in the absence of disease is the physiological basis of efficient vaccination1. Recently, systems biology-based approaches have revealed that successful vaccines such as the yellow fever vaccine, induce a strong induction of innate immune responses and activation of several subsets of dendritic cells (DCs), which in turn, lead to multilineage activation of antigen-specific T cells2,3. Since DCs are the only immune cell population with the ability to activate antigen-specific naïve T cells4, the study of their function during vaccination is critical to understand immune responses to vaccines and to design future strategies against challenging pathogens.
A system allowing tracing of different DCs subsets during immune responses to vaccines would be desirable in order to establish an accurate kinetics of DC migration to lymphoid tissues, and therefore to provide insight into the physiological mechanisms responsible for the initiation of vaccine-specific adaptive immunity. Reverse genetics-based approaches offer the possibility to generate modified, live-attenuated vaccines that can be used experimentally with this purpose. Since its implementation on influenza research, plasmid-based reverse genetics has been widely employed to generate recombinant influenza strains including LAIVs. Standard protocols to rescue recombinant influenza viruses require multi-transfection of highly transfectable cell lines with ambisense plasmids (producing both positive and negative sense RNA) containing the eight influenza viral segments as well as amplification in a permissive system such as Madin-Darby canine kidney (MDCK) cells and/or chicken embryonated eggs5. However, the application of reverse genetics to generate molecular tools in order to study the immune mechanisms of vaccination remains unexplored.
The generation of new mouse models allowing specific depletion of immune cell subsets, including DCs, has opened new possibilities to understand the basic immune mechanisms underlying vaccine-elicited protection. The comparison between DC subset functions in mice and humans has revealed that, to a great extent, mouse and human DCs are functionally homologous6,7, these findings, strongly suggest that the development of mouse models allowing specific depletion of DCs in the steady state and during inflammatory conditions, may serve to understand the physiology of DC responses in humans. In recent years a number of mouse models have been generated carrying transgenes expressing the simian diphtheria toxin (DT) receptor (DTR) under the control of the promoter region of a gene of interest8,9. Since mouse tissues do not naturally express DTR, these models allow conditional depletion of cell subsets carrying the targeted gene of interest upon mouse inoculation with DT. Thus, our ability to deplete specific DCs and other leukocytes in vivo during physiological processes, has been greatly enhanced by the development of DTR-based ro. However, while these transgenic mouse models have been used extensively to understand the ontogeny of the immune system, their application to vaccine development has been scarcely tested. Here, by combining influenza reverse genetics and DTR-based competitive bone marrow chimeras, we propose a method to study the kinetics of vaccine immunity as well as individual gene function during immune responses to vaccines in vivo. The application of this technology for preclinical evaluation of new vaccines against challenging infectious diseases could help to rationalize vaccine design and to test vaccine candidates in vivo.
Animal experiments were conducted according to approved protocols and following the guidelines of the German animal protection law. All staff carrying out animal experiments passed training programs according to category B or C of the Federation of European Laboratory Animal Science Associations.
1. Generation of Recombinant Live Attenuated Influenza Vaccines by Reverse Genetics
NOTE: The detailed protocol for the generation of recombinant influenza viruses by reverse genetics has been described by previous studies5 and is out of the scope of this report. Briefly, rescue of cold-adapted influenza vaccines (CAV) is done in a biosafety class II cabinet under biosafety level 2 (BSL2) containment, and involves steps detailed below. The transfection and infection protocol follows that of Martinez-Sobrido et al.5.
2. Tracking Vaccine-specific Immunity
3. Chimeric Mice to Dissect Gene-specific Vaccine Responses
NOTE: The generation of competitive bone marrow chimeras using DTR-based mouse models allows the discrimination of cell-specific functions during immune response to vaccines. Here we show an additional strategy by which combination of DTR-expressing donor cells with cells from knockout mice permits to study gene specific functions during immunity in specific cell compartments. In the example we generate mice with specific deletion of toll-like receptor 3 (TLR3) in Langerin+ CD103+ DCs.
Generation of recombinant live-attenuated influenza vaccines can be achieved by transfection of plasmids encoding the eight segments of influenza virus under the control of bidirectional promoters5. A cold-adapted influenza vaccine usually contains six segments of a cold-adapted strain as well as the HA and NA of the influenza strain of choice (e.g., H1N1) (Figure 1A). The principle of cold-adaptation is based on virus-restricted replication at 33 °C, the temperature of the upper-respiratory tract (URT) in mice and humans14. Therefore, the vaccine can replicate to some extent in the URT but cannot cause lower respiratory tract pneumonia. Using fusion PCR technology, a conserved region in the stem of the viral neuraminidase (NA) was substituted for a traceable OVA-derived peptide (SIINFEKL) (Figure 1B).
To track vaccine-specific responses in kinetics studies we have taken advantage of traceable peptides derived from the vaccine formulation as well as chimeric mouse models. Between day 3 and 6 post-vaccination, SIINFEKL-bearing CD103+ DCs can be detected in the lung-draining lymph nodes12 (Figure 2A). Multiparametric flow cytometry allows quantification of vaccine-derived antigen presentation in real time. In addition, infusion of SIINFEKL-specific T cells allows quantification of vaccine-specific CD8 T cell responses (Figure 2B).
To evaluate gene-specific functions during vaccination we devised a mouse model combining DTR technology and competitive bone marrow chimeras (Figure 3A). By doing so, loss of gene function was targeted to specific cell compartments over the course of immune responses to vaccination (Figure 3B, C).
Figure 1. Engineering of traceable LAIVs. (A) Influenza segments cloned in ambisense plasmids (pDZ backbone as described by Martinez-Sobrido et al.5) were transfected into 293T cells using Lipofectamine 2000. 24 hr after transfection, supernatants obtained from 293T cells were used to coat influenza-permissive Madin-Darby canine kidney (MDCK) cells in the presence of 1% of TPCK trypsin. All the protocol was performed at 33 °C. Viral supernatants from MDCK cells were then inoculated into 9-day-old embryonated chicken eggs for three days at 33 °C. Viral rescue was confirmed by PCR-based viral genome amplification and sequencing. (B) To insert the OVA-derived SIINFEKL peptide into the influenza neuraminidase (NA), a conserved region (residues 65 to 72) of the NA gene of A/Puerto Rico/8/34 (H1N1) virus was substituted for a short cDNA encoding the SIINFEKL peptide via fusion PCR. Please click here to view a larger version of this figure.
Figure 2. Vaccine tracing techniques. (A) Migratory DCs were gated in the mediastinal lymph nodes (mLNs) as CD11cmed MHC class IIhi cells. In the example, antigen-bearing migratory CD103+ DCs were identified as CD103+ H-2b-SIINFEKL+ cells by flow cytometry. (B) OT-I-specific T cells were infused to mice on the same day of vaccination. Four days after, cells were collected from the lymph nodes of vaccinated mice and assessed for their proliferation profile. CFSE dilution on the FL-1 channel was utilized as indicator of cell division. CAV: Cold-adapted vaccine; CAVSIINFEKL: Cold-adapted vaccine expressing OVA-derived SIINFEKL peptide. Please click here to view a larger version of this figure.
Figure 3. Gene function studies. (A) Schematic of bone marrow chimeras (BMC) and vaccination strategies. (-2 and +2 refer to administration of DT two days before or after vaccination). (B) Addition of DT in chimeric mice expressing Langerin-DTR allows specific depletion of migratory Langerin+ DCs coexpressing CD103 in the lung. (C) Comparative analysis of vaccine-specific T cells in chimeric mice. Detection of NP366-374-specific CD8 T cells was done via commercial dextramer staining. Please click here to view a larger version of this figure.
In this study we describe how reverse genetics and chimeric mouse models can be utilized to elucidate the physiologic and molecular mechanisms of vaccine-induced immunity. Influenza reverse genetics is established in many laboratories and has played a chief role in understanding influenza pathogenesis, replication, and transmission17. A key point in our protocol is rescue of cold-adapted influenza vaccines expressing foreign epitopes. While the strategy of introducing short cDNAs into the stalk of the neuraminidase has been described by many groups, the investigator needs to ensure that no additional mutations are introduced during vaccine production in eggs and that the vaccine is able to replicate in the respiratory tract without inducing pneumonia5,12. Since both the stalk of the viral neuraminidase as well as other regions of the influenza genome such as the NS gene segment and the PB1 segment allow allocation of foreign sequences19,20, modifications of our protocol can be utilized to add additional foreign peptides to, for example, understand the hierarchies of the T cell immune response against influenza virus, a key point for rational vaccine design.
By combining traceable vaccines and DTR-based depletion of DC subsets we have been able to target loss of gene function to specific cell subsets. This technology has been applied before to elucidate immune response to pathogens21, but not to dissect pathways responsible for vaccine protection. Due to the availability of DTR-based models and the flexibility of the technique to generate chimeric mice, this technique could be further expanded to additional leukocyte populations and genes with immune functions. It is important to note that DT treatment may also result in depletion of additional DC subsets expressing langerin such as CD8α+ DCs in the lymphoid tissues. Therefore it is highly recommended to perform a dose-response study to evaluate the effects of DT treatment. A potential caveat of our model is that allogenic bone marrow transplantation has been shown to decrease overall antiviral CD8 T cell immunity15. Thus, caution has to be exercised when comparing data between transplanted and non-transplanted mice.
Several key human pathogens for which vaccines are urgently needed can be studied in mouse models using wild-type pathogens or mouse surrogates. These include influenza virus, Plasmodium, and Ebola virus. The techniques proposed in this study may help to understand the basic mechanism of experimental vaccines against these pathogens and therefore, to rationalize vaccine design.
The authors have nothing to disclose.
We thank Sergio Gómez-Medina for excellent technical support with mouse experiments. This work was supported by funds from the Leibniz Association and the Leibniz Center of Infection. A.L. is a recipient of a pre-doctoral fellowship from the Leibniz Graduate School.
Dulbecco´s Modified Eagle Medium (DMEM 1X) | Gibco RL-Life Technologies | 41965-039 | |
Opti MEM | Gibco RL-Life Technologies | 31985-047 | |
Lipofectamine 2000 | Invitrogen-Life Technologies | 11668-027 | |
Penicillin-Streptomycin (10.000 U/ml) | PAA | p11-010 | |
Bovine Serum Albumin | Sigma-Aldrich | A2153 | |
Embryonated eggs | Valo biomedia Gmbh | ||
PBS (1X) | Sigma-Aldrich | D8537 | |
70 μM Nylon Filters | Greiner-Biorad | 542-070 | |
Red Blood Cell Lysing buffer (RBCL) 10X | BD Bioscience | 555899 | |
CD16/CD32 Mouse BD Fc Block (2.4G2) | BD Pharmigen | 553142 | |
APC-Anti-mouse SIINFEKL-H2kb (25 D1.16) | Biolegend | 141605 | |
PE-Anti-mouse CD11c (HLA3) | BD Biosciences | 553802 | |
eFluor 450-Anti-mouse MHCII (Md/114.15.2) | eBioscience | 48-5321-82 | |
Pe-Cy7-Anti-mouse CD11b (M1/70) | Biolegend | 101216 | |
PerCp/Cy5.5-Anti-mouse CD103 (2E7) | Biolegend | 121416 | |
PE-Anti-mouse CD45.1 (A20) | eBioscience | 12-0453-82 | |
V500-Anti-mouse CD45.2 (1O4) | BD Bioscience | 562130 | |
PerCp-eFluor710 -Anti-mouse CD8a (53-6.7) | eBioscience | 46-0081-80 | |
APC-Cy7-Anti-mouse CD3ε (145-2611) | Biolegend | 100325 | |
eFluor450-Anti-mouse CD4 (GK 1.5) | eBioscience | 48-0041-80 | |
CFSE Proliferation dye | eBioscience | 65-0850-85 | |
Baytril 2.5% | Bayer | 65-0850-85 | |
Dymethil-Sulfoxide (DMSO) | Sigma-Aldrich | D2650 | |
Ovalbumin | Molecular probes | O23020 | |
Diphteria Toxin (DT) | Sigma-Aldrich | D0564 | |
Trypsin-TPCK | Sigma-Aldrich | T1426 | |
BD FACsCanto II Flow cytometer | BD Biosciences | ||
FlowJo cell analysis software 9.5 | Flowjo inc. | ||
Trypan Blue Stain (0.4%) | Life technologies | T10282 | |
Countess Automatic Cell Counter | Invitrogen-Life Technologies | C10227 |