Influenza Immunity: Protective Mechanisms Against Pandemic
Respiratory Virus

In this renewal application entitled "Influenza Immunity: Protective Mechanisms Against Pandemic Respiratory Virus," we intend to build on the progress we have made in the previous granting period with two overall objectives in mind: the first is to extend our knowledge of the human immune response to influenza vaccine much more broadly and deeply across different age groups and with different vaccine modalities and to probe the influence of genetics on these responses using monozygotic and dizygotic twin pairs. A second and new theme that we are pursuing is to use our analyses of responses to the two licensed influenza vaccines to establish a unique dataset and "metrics" of human immunology in which we can begin to identify biomarkers for what is a healthy and effective response versus what is not. In our reformulated team, we have significant strengths in immunology, infectious diseases, vaccinology, molecular biology, immune monitoring, bioengineering, genetics and bioinformatics. We also have an outstanding clinical core that has been extremely good at recruiting and processing subjects of all ages into our vaccine studies. Enhancing our clinical cohort as well as a new relationship with SRI International, which has a large and expanding twin registry in innate and adaptive immunity, the members of which will form the core of our subject base. In order to extract the maximum amount of information from each clinical specimen, we have added two new cores to our infrastructure: the Stanford Human Immune Monitoring Core to provide an in depth and uniform immunological analysis of each subject, and a Bioinformatics Core to organize the large quantities of data and to help investigators interpret it. In our projects, we focus on B and T cell responses and repertoire as well as biomarkers and signaling pathways, and in our Technology Development projects, we seek to push the boundaries of repertoire and single cell analysis and also to develop innovative new bioinformatics tools that will be vital to understanding the complexities of the immune system and the response to the influenza virus.

A formidable and virtually unique challenge in influenza vaccine development is that the human population is repeatedly exposed to changing influenza viruses, which render previously immune individuals vulnerable to newly emerged strains, thus causing annual epidemics or even pandemics of influenza. While the human immune system has the intrinsic capability of coping with highly diversified viral antigens with its enormous antibody repertoire (estimated at a magnitude raised to the 11th power), it is our hypothesis that this potential diversity is significantly restricted in certain age groups, such as the elderly and the very young, and is not fully realized following immunization with certain types of vaccine in some age groups, such as the trivalent inactivated influenza vaccine (TIV) compared to the live attenuated influenza vaccine (LAIV) in young children. In this application we propose to address these issues using both serologic and molecular approaches, with an emphasis on the age groups most vulnerable to influenza morbidity and mortality - very young children and the elderly. The specific aims of the proposal are: A1. To compare the antibody responses against the vaccine strains (homotypic reactivity) and mismatched virus strains (heterovariant reactivity) after immunization with LAIV or TIV. We will characterize and compare the heterovariant antibody responses in a collection of paired serum samples from young children, healthy adults and elderly who are vaccinated with these two types of influenza vaccines. We will also clone and express immunoglobulin (Ig) genes from individual antibody-secreting cells (ASCs) elicited shortly after immunization with TIV or LAIV in the different age groups and compare the specificity and affinity of these monoclonal antibodies against both homotypic and heterovariant influenza viruses. A2. To identify factors affecting the sequences of Ig genes encoding influenza-specific antibodies and determine the relationship between specific Ig gene sequence usage and function. We will carry out a systematic analysis of the sequences of Ig genes isolated from ASCs after immunization with LAIV or TIV, and relate the sequence characteristics of Ig genes to antibody reactivity against different influenza strains. In particular, we will compare the Ig gene sequences between the elderly, younger adults and children, between the recipeints of LAIV and TIV and between LgG and LgA isotypes.

In this Project 2 entitled "T cell and General Immune Responses to Influenza," we focus on general immunological biomarkers and antigen specific T cells in a continuing effort to define the immune response to influenza vaccination in volunteers from different age groups and especially utilizing the twin cohorts of SRI.  We hope to establish benchmarks of immune proficiency in these different cohorts that will be predictive not just of an influenza response, but of immunological "health," or at least a component of it, in general.

In Aim 1, we will continue our ongoing program to survey the blood samples of volunteers being immunized for a broad array of serum cytokines, white blood cell subsets, lymphocyte proliferation, and whole genome gene expression. Using the twin cohorts, we will be able to ask what traits are determined genetically, and of those evident in young adults, which change in older adults. Also a high priority is what traits correlate with a robust or poor response.

In Aim 2 and 3, we will take advantage of recent advances in peptide-MHC tetramer technology together with our own efforts to probe more broadly and deeply into the nature of the T cell response to different influenza antigens. Thus, we will survey dozens of different T cell epitopes at once to ask whether or not there are specific changes in influenza specific T cell repertoires with age and to what extent does genetics contribute to the repertoire. We will also ask how the different influenza vaccine types influence the repertoire of responding T cells and whether or not there is a correlation between the repertoire used, and a robust or poor response. We have also been able recently to isolate antigen specific naive populations of T cells (which are in the cytokines produced in the signaling pathways utilized; in some cases, only 1 in one million cells of the CD8+ T cell pool). This gives us the ability to probe the naive repertoire of strain-specific T cells in young adults and characterize their response to different types of influenza vaccination.

Lastly, in Aim 4, we propose to use "humanized"mice to test hypotheses generated in the first three aims.

Intracellular protein levels, subcellular localization, or activation state are reflective of a cell’s functions. Some relevant cell populations are so rare as to make their isolation for standard biochemical analysis essentially impossible.  We have previously shown that disease-driven, single-cell intrinsic events can have profound effects on phospho-signaling network architecture and this can be correlated to clinical outcomes.  In the case of viral infection, it is unclear whether or not certain individuals possess immune characteristics that make them more or less susceptible to influenza infection. For instance, environmental or individual characteristics such as age and immune system health could have effects upon the immune system's response to viral challenge.

We will document the signaling biology of the immune system at two levels of resolution.  First, we will investigate and document the changes to immune signaling post-influenza infection of human PBMC in vitro.  These studies allow for analysis of influenza infection and the changes that it creates in immune cell subsets at a single cell level. Second, we will study the signaling biology of PBMC in age-selected cohorts of healthy subjects, including monozygotic twins, given either of the two licensed influenza vaccines (TIV or LAIV).  These studies provide a fuller understanding of how immune system changes in the young, the healthy adult, and the elderly individual might account for differing response patterns to alternative vaccine strategies and provide in-sights about influenza effects upon signaling behaviors of immune system cells.

Measuring the Immunome: Genomic Approaches to B Cell Repertoire. In this project, we propose to develop new technological applications of next generation DNA sequencing, high throughput PCR chips, and microfluidic single cell processors to perform systemic analyses of the human B cellrepertoire. Using B cells purified from human peripheral blood, we will apply these technologies to try to undersand the effects of three phenomena on immune repertoire: aging, genetic background and vaccination history. Four groups of subjects will be studied: 1. age groups - children, young adults and elderly; 2. genetic background - identical twins vs. fraternal twins vs. not-related subjects; 3. vaccination history - pre, post and multiple immunizations; 4. vaccination type- the trivalent inactivated influenza vaccine (TIV) and the live attenuated influenza vaccine (LAIV). This study will provide the first comprehensive and systematic measurements of B cell repertoire at three different anatomic levels, V/D/J/C exon usage, immunoglobulin (Ig) gene deep sequencing, and single cell gene expression analysis. Our specific aims are:

• Aim 1. Develop a quick and quantitative assay to measure the V, D, J, and C exon usage landscape.
• Aim 2. Develop a high throughput sequencing protocol to perform dep sequencing of Ig heavy chain gene diversity.
• Aim 3. Develop a microfluidic parallel processor for single cell analysis of B cell repertoire.
• Aim 4. Dissect the influences of aging, genetic background, and vaccination history on the human B cell repertoire.

The immune system, by any standard is complex. As the number of components in a system and interactions between components grows, our ability to fully comprehend the system diminishes significantly. Principally, humans in comparison to computers, do poorly in tasks such as memorizing, searching and constructing multi-component, multi-scale images. The need to "free our knowledge" from free text, and place it in a computer system which will memorize our knowledge, allow for easy searching and visualization of our understanding of the immune system is apparent. The use of natural language processing to represent molecular findings from the scientific literature, and the use of network visualization have both made a clear impact across biology. Here, we plan to build a new informatics tool that uses both of these techniques to represent the complex network between cells, immunological processes, secreted cytokines, expressed transcripts, and phosphorylated signaling proteins, called ImmuneXpresso. We will then apply this tool to the multiple type of molecular measurements made throughout this program, to suggest the cellular cause of influenza vaccine non-response. Our tools will be available for beta tested to investigators within this program and released to the general community. 

The objective of this Cooperative Center for Translational Research on HumanImmunology and Biodefense is to use analysis of vaccine-induced and naturally acquired immunity to influenza A as a model for defining adaptive and innate immune mechanisms and antiviral protection in children and adults. The Clinical Research Core will be responsible for coordinating protocol design and implementation, providing biostatistical support, obtaining human subjects approvals, and creating and managing the centralized database to record clinical and laboratory data. The Clinical Reserach Core will include a laboratory toreceive blood and respiratory specimens, carry out initial sample processing and to distribute relevantspecimen to the participating laboratories, perform serology assayas and analyses. Centralizing thesefunctions is particularly important to assure the most efficient use of small volume pediatric blood samples. As the work proceeds, the Core database will facilitate comparative analyses of results obtained from the individual Research projects. The Specific Aims are:

• Aim 1: to design the clinical study and data analysis plan, make the necessary IRB submissions for clinical studies, recruit and enroll adults and children into clinical protocols, assure that subject rights are respected, and provide follow-up to assure collection of complete sets of data.
• Aim 2: to provide initial sample processing and distribution of blood or salivasamples to the Principal Investigators, perform standard serologic assays to determine baseline influenza A immune status and measure antibody responses to vaccination or natural infection. The Core will also perform direct influenza A rapid diagnostic tests to recruit children for the study of natural influenza A infection. Respiratory samples for a subset of children in the vaccine study will be collected weekly by Core personnel during the flu season.
• Aim 3: to provide centralized data management and biostatistical supportfor the Center. The Clinical Reserach Support Core will thus administer the clinical protocols under which samples are collected for use in all Research Projects and Research Resource Technical DevelopmentProjects. 

The role of the Human Immune Monitoring Core (HIMC) within this Cooperative Center for Translational Research on Human Immunology and Biodefense is to serve as a central facility for collecting, analyzing and storing clinical samples in order to maximize opportunities for parallel evaluations across all projects, running basic immune assays on our Luminex, flow cytometry and gene array platforms, and creating a steady flow of deidentified sample and experimental results into the Bioinformatics core that will then be organized and available to all project PI's through that database. The HIMC includes a laboratory to receive processed samples from the clinical core and will also serve as a repository for materials needed for future studies by the project PI's. The Specific Aims of the HIMC are: 

• Specific Aim 1: Specialized Immune Monitoring assays. This Scienc eCore will carry out of 42 plex Luminexassays on the serum to assess immune response of the patients at the time fo blood draw, Agilent arrays to assess gene expression in whole blood and flow cytometry to phenotype leukocyte populations and to assess the proliferative response to specific cytokine and mitogen stimulation.
• Specific Aim 2: Centralized Assay Results database and analysis support. The Science Core will develop a repository for all patient samples run that will be linked to all archived samples, all data generated by the assays and all assays performed on data transferred to other projects with a unique barcode identifier. Thisensures that all data and samples handled by the Core are deidentified in keeping with HPPAA regulations. 

The purpose of the Bioinformatics Core is to develop, validate, and use several established bioinformatics tools for the study of protective mechanisms against pandemic respiratory virus. The informatics platforms developed through the Bioinformatics Core will be used to study samples provided by Projects 1-3. Datasets, protocols, reagent lists, and informatics tools will be placed on a website that will be developed as part of this proposal. The goals of the Bioinformatics Core will be

(1) to connect with all the various Projects that are generating primary data, 

(2) to acquire all the relevant data and store it in open formats, 

(3) to collaborate with the other NIH-funded Cooperative Centers for Translational Research on Human Immunology and Biodefense and share data in a bi-directional manner, 

(4) to serve data to the Projects for hypothesis testing, 

(5) to publish the data for public availability, and 

(6) to provide robust statistical and analytical methods to analyze the data. 

The highest priority for the Bioinformatics Core is to directly work with all Projects to address their need for robust statistical techniques. In addition to analytic support, the Bioinformatics Core will operationalize collaboration, data-, and method-sharing with other NIH-funded CooperativeCenter for Translational Research on Human Immunology and Biodefense. Finally, the Bioinformatics Core will work with all Projects to publish data to the Internet. To achieve these goals, the Bioinformatics Core will create a software infrastructure to enable state-of-the-art distribution, storage and analysis of multiple types of genome-scale data. This will enable researchers from all Cooperative Centers to maximally utilize the genomic, proteomic, immunogenomic, and phenotypic data sets to determine functional dependencies among the measured elements and direct further biological validation of these putative dependencies.

The protozoan parasite Toxoplasma gondii is a Category B Biodefense Pathogen as defined by NIAID.  This obligate intracellular agent causes a spectrum of human disease and recent data suggest that this variability is partly dependent on which strain is responsible for the infection.  Our work and that of others have shown that a key part of strain-specific differences in virulence in animal models is due to polylmorphic protein kinases that are secreted by Toxoplasma.  These secreted effectors have markedly different effects on the immune response in mice, depending on the allele carried by the parasite.  In vitro analyses of human cells also show dramatic strain-specific effects on cytokine expression.  Our hypothesis is that these strain-specific effectors are also key to the outcome of infection in people.  To test this hypothesis, we will examine a cohort of people that are infected with Toxoplasma and assess their disease symptoms, the strain of parasite responsible for the infection and, through use of the Human Immune Monitoring Core, the immune profile during acute and chronic infection.  Correlations between these three variables will provide not only a much better understanding of the parasite’s pathogenesis but also the means to readily predict disease outcome and, thus, appropriately manage patients in a given outbreak. 

Each year, influenza virus infects between 5 to 15% of the population, causing millions of cases of severe respiratory infection and hundreds of thousands of deaths worldwide. Current vaccines provide limited immunity to yearly circulating strains, but not broader protection spanning multiple years or multiple subtypes of the virus. Our understanding of the details of the neutralizing B cell response to flu infection and vaccination remains incomplete, yet deeper insight into this response might be very helpful in guiding the development and evaluation of new influenza vaccine approaches. We propose to use fragments of a major antibody target on the surface of the virus, the influenza hemagglutinin, to develop novel tools for mapping the locations and diversity of neutralizing B cell epitopes. These novel reagents should provide new insights into the human B cell response to flu as well as aid future vaccine development.

We have identified a non-coding RNA, termed NeST, as a positive regulator of interferon gamma (IFN-γ) expession in mice. In transgenic animals, the expression of this 934-nucleotide noncoding RNA is sufficient to lessen disease severity during experimental Salmonella infection and to increase the abundance of interferon gamma (IFN-γ) in cultured CD8+ cells following ex vivo stimulation. Humans express an RNA species homologous to NeST, which also appears to be noncoding and transcribed adjacent to the IFN-γ locus, from the opposite DNA strand. Furthermore, in humans, a polymorphism that correlates with inflammatory disease has been ascribed to the first intron of the IFN- γ-encoding gene: however, this same location also corresponds to the fifth intron of the overlapping NeST RNA-encoding gene.

We will determine whether the expression of NeST RNA correlates with the expression of IFN-γ in immune cells from several different individuals, both healthy people and those suffering from rheumatoid arthritis. We will use RNAi knock-down and cDNA-based expression interventions to introduce perturbations into these primary cells to observe any effects on the expression of IFN-γ and other cytokines. Recent findings of our Stanford colleagues have revealed dramatic subgrouping in the rheumatoid arthritis population: those that express the IFN-γ at high abundance and those that do not. In pilot experiments to test the diversity of NeST expression among people, we have chosen this phenotypically diverse group for analysis. Understanding individual variation in both coding and noncoding genes of the human immune response is crucial to the development of a personalized approach to medicine. 

Michael Hsieh

Approximately half of all girls and women will have at least one urinary tract infection (UTI) in their lifetimes. In girls and women prone to recurrent UTI, the vagina may serve as a reservoir for bacteria such as uropathogenic E. coli (UPEC). Another common pathogen which infects the vagina is herpes simplex virus(HSV)-2. Both UPEC and HSV-2 infections may be kept in check through CD1d, a non-polymorphic molecule which presents microbial lipid antigens to invariant NKT cells2. Activation of invariant NKT cells by CD1d-lipid complexes triggers an antimicrobial immune response. Baseline CD1d expression is particularly high in human vaginal epithelium, suggesting this molecule plays a critical role in vaginal immunity3. Indeed, administration of synthetic CD1d ligands to female mice protects them against UTI4 and otherwise lethal genital HSV-2 infection5. Hence, induction of CD1d expression and signaling may be a novel immunostimulatory approach to prevent and treat UTI and genital HSV-2 infection in humans6. Before CD1d-based therapeutics can be pursued in earnest, improved knowledge of the relationship between CD1d and commensal bacteria is needed. Indeed, interference with normal CD1d functions disrupts commensal bacteria in the mouse gut. Conversely, the presence of commensal bacteria in the mouse gastrointestinal tract has been shown to regulate development of CD1d-restricted NKT cells in multiple publications8-12. However, the role of vaginal commensal lactobacilli in regulating CD1d functions is unknown. IFN-γ is the only reported inducer of CD1d expression in human vaginal epithelium3. Although CD1d exhibits activity against UPEC and HSV-2, other vaginal pathogens actively thwart CD1d signaling. Human papillomavirus (HPV) E5 targets human vaginal CD1d to the cytosolic proteolytic pathway by inhibiting calnexin-related CD1d trafficking 13. Thus, decreased CD1d expression in the presence of HPV E5 may help HPV-infected cells evade protective immunological surveillance. This is the only report of the molecular pathways that downregulate human vaginal CD1d expression. To address this dearth of knowledge regarding CD1d in the human vagina, we have established human vaginal epithelial cell-bacterial co-culture techniques. These approaches facilitate the study of human vaginal innate immune responses to microorganisms. Specifically, we have found that vaginal epithelial CD1d expression is increased by co-culture with UPEC. Moreover, UPEC-induced CD1d upregulation is abrogated when commensal lactobacilli are added to vaginal epithelial cell-UPEC co-cultures. We hypothesize that: 

1. discrete UPEC proteins upregulate CD1d promoter activity by human vaginal epithelial cells, and 
2. Lactobacillus proteins interfere with the ability of UPEC to increase CD1d promoter activity in human vaginal epithelial cells. We will use transposon libraries and genome-wide, high throughput screening methods to test this hypothesis, which will advance human immunology by further defining human vaginal CD1d biology. This work will pave the way for CD1d-based therapeutics for UTI and HSV-2 infection.