Can Systems Biology Improve Influenza Immunization Practice?
By Amesh A. Adalja, MD, FACP, July 22, 2011
Each influenza season, a substantial proportion of people who are immunized, and especially the elderly, fail to mount a protective response to the vaccine. A protective response is defined as a 4-fold increase in the IgG antibody titer targeted to the viral hemagglutinin protein.
With a systems biology approach,* the changes that occur in gene expression and metabolic pathways after vaccination can be studied to identify specific alterations that predict response to vaccination. This approach could also yield clues to improving vaccines by enhancing effectiveness and minimizing adverse events. Such studies have been conducted with the live-attenuated Yellow Fever vaccine but not with inactivated vaccine, such as the trivalent influenza vaccine (TIV).
Nakaya and colleagues have just published results of a series of experiments that used systems biology to predict human immune response to TIV and to uncover potential explanations for the vaccine’s poor immunogenicity.2
The researchers administered TIV to 28 healthy young adults during the 2008 influenza season. Subjects were classified as either high or low responders based on their 28-day antibody titers. This response was only modestly correlated (r=0.43) with the number of IgG-secreting plasmablasts, (antibody secreting B-cells in the blood) highlighting the need for a more definitive correlate of protection.2
Molecular Signature of an Influenza Vaccine
Vacinees’ blood was surveyed on days 0, 3, and 7 for molecules known to play a key role in immune response. Of 9 candidate molecules, 1 was significantly altered by TIV. CXLCL10 (IP-10, interferon gamma-induced protein 10) was found to demonstrate a modest negative correlation to 28-day antibody titer (r = -0.48). As with levels of IgG-secreting plasmablasts, levels of CXLCL10, a chemoattractant for immune cells,3 were only modestly correlated with protection, so Nakaya and colleagues searched for a better correlate of protection.2
Further studies included microarray analysis of the transcriptome of peripheral blood mononuclear cells (PBMCs) collected on days 0, 3, and 7 post-vaccination. Important findings included upregulation of antibody part encoding genes and protein unfolding response genes, due to the large amount of misfolded proteins that accumulate in the endoplasmic reticulum of antibody secreting cells.2
Network analysis of patterns of gene expression also identified links among innate immunity, cell-mediated immunity, and antibody response.2
Finding Molecular Signatures to Predict Antibody Response
Analysis of genetic signatures was further parsed to find the minimum set of genes needed to predict antibody response to TIV. Using a subset of TIV responders who exhibited an extreme response (8-fold or higher increase in titer) to TIV, 44 genes were found that could predict very low or very high response with 85% to 90% accuracy. When the analysis was performed with the conventional response threshold of a 4-fold increase in titer, 85% predictability was achieved with a 42-gene signature model.2
A Novel Hypothesis: The Role of CaMKIV
In order to show that new facets of vaccine response could be uncovered using a systems biology approach, one gene from the molecular signature was selected to confirm the findings—CaMKIV, a gene not known to be involved with B-cell response.2
In further study, CaMKIV expression demonstrated a negative correlation with antibody response and plasmablast population size after vaccination. In a mouse splenocyte model and human PBMCs, TIV administration caused the phosphorylation of CaMKIV, suggesting that vaccination activates CaMKIV. Further experiments on CaMKIV-deficient mice revealed higher antibody titers after TIV administration when compared with wild-type mice.2 The authors concluded that CaMKIV is important in regulating the B-cell response to influenza vaccination, and their methodology validated the prediction made by their systems biology study approach.
Unlocking the Mysteries of Immune Response
Systems biology has the potential to elucidate novel pathways that have escaped identification by conventional study methods. Understanding the changes in the host that occur after vaccination and infection by a wild-type pathogen will improve not just vaccination practice, but diagnostic and prognostic capabilities as well. With a vaccination, it may become feasible to obtain follow-up blood samples a few days after vaccine is administered and assess whether an appropriate response has occurred or whether boosters or an alternative vaccine (adjuvanted, subcutaneous, high dose, live vs. inactivated, etc.) should be administered. Systems biology can also be applied to test the efficacy of vaccines in different stages of development.
In the realm of diagnosis, identification of specific host gene signatures in response to pathogens at both the kingdom level (bacteria, virus, fungus, protozoa, helminth) and more detailed levels (gram negative vs. gram positive bacteria, genus, and species) may help delineate the cause of illness faster than conventional techniques. Additionally, systems biology has the potential to identify a pattern of gene expression alteration that may correlate with good or poor prognosis, which would allow physicians to match the intensity of intervention with the most likely outcome of illness.
*Systems biology has been defined as “the study of an organism, viewed as an integrated and interacting network of genes, proteins and biochemical reactions which give rise to life.”1
Systems biology: the 21st century science. Institute for Systems Biology. http://www.systemsbiology.org/Intro_to_Systems_Biology/Systems_Biology_--_the_21st_Century_Science. Accessed July 13, 2011.
Nakaya HI, Wrammert J, Lee EK, et al. Systems biology of vaccination for seasonal influenza in humans. Nat Immun 2011; doi:10.1038/ni.2067.
Dufour JH, Dziejman M, Liu MT, et al. IFN-gamma-inducible protein 10 (IP-10; CXCL10)-deficient mice reveal a role for IP-10 in effector T cell generation and trafficking. J. Immunol 2002. 168:3195–204.