Into the eye of the cytokine storm

Abstract

The cytokine storm has captured the attention of the public and the scientific community alike, and while the general notion of an excessive or uncontrolled release of proinflammatory cytokines is well known, the concept of a cytokine storm and the biological consequences of cytokine overproduction are not clearly defined. Cytokine storms are associated with a wide variety of infectious and noninfectious diseases. The term was popularized largely in the context of avian H5N1 influenza virus infection, bringing the term into popular media. In this review, we focus on the cytokine storm in the context of virus infection, and we highlight how high-throughput genomic methods are revealing the importance of the kinetics of cytokine gene expression and the remarkable degree of redundancy and overlap in cytokine signaling. We also address evidence for and against the role of the cytokine storm in the pathology of clinical and infectious disease and discuss why it has been so difficult to use knowledge of the cytokine storm and immunomodulatory therapies to improve the clinical outcomes for patients with severe acute infections.

Fig 1

Imagery of a cytokine storm.

Fig 2

Human ARDS. Photomicrographs from the lungs of 2 different patients with ARDS, stained with hematoxylin and eosin (H&E) are shown. The alveolar spaces are filled with a mixed mononuclear/neutrophilic infiltrate, the alveolar walls are thickened, and the septae are edematous. Note the presence of cellular debris and proteinaceous material in the air spaces (A [magnification, ×200] and B [magnification, ×400]). In later stages (C [magnification, ×200] and D [magnification, ×400]), there is a fibroproliferative response with collagen deposition in the alveolar walls (arrows). Note that the alveolar epithelium has been replaced with cuboidal cells (arrowheads). (Figure and legend reprinted from reference with permission of the publisher.)

Fig 3

H5N1 and H1N1 influenza viruses differentially regulate the acute-phase response during infection. The diagrams show transcriptional changes for cytokine signaling pathway molecules in mouse lungs infected with either H5N1 virus (27) (A), 1918 virus (27) (B), or seasonal H1N1 virus (73) (C). Microarray data were background corrected and normalized by LOESS and quantile normalization. Expression values were represented as log₂ ratios of infected to respective mock-infected samples, where biological and technical replicates for a given condition were averaged. PathVisio was used for data visualization (143) of the acute-phase response signaling pathway from IPA (Ingenuity Systems). Official gene symbols are shown. Serum amyloid (Saa2 and Saa3) and serine (or cysteine) peptidase inhibitor (Serpina1d and Serpina3a) represent acute-phase response proteins. Red indicates that the gene expression is increased relative to that for the uninfected reference. Green indicates that the gene expression is lower than that for the uninfected reference. White indicates no change in gene expression, and gray indicates that a molecule was not detected by microarray analysis. inf, infected; d, days postinfection.

Fig 3

H5N1 and H1N1 influenza viruses differentially regulate the acute-phase response during infection. The diagrams show transcriptional changes for cytokine signaling pathway molecules in mouse lungs infected with either H5N1 virus (27) (A), 1918 virus (27) (B), or seasonal H1N1 virus (73) (C). Microarray data were background corrected and normalized by LOESS and quantile normalization. Expression values were represented as log₂ ratios of infected to respective mock-infected samples, where biological and technical replicates for a given condition were averaged. PathVisio was used for data visualization (143) of the acute-phase response signaling pathway from IPA (Ingenuity Systems). Official gene symbols are shown. Serum amyloid (Saa2 and Saa3) and serine (or cysteine) peptidase inhibitor (Serpina1d and Serpina3a) represent acute-phase response proteins. Red indicates that the gene expression is increased relative to that for the uninfected reference. Green indicates that the gene expression is lower than that for the uninfected reference. White indicates no change in gene expression, and gray indicates that a molecule was not detected by microarray analysis. inf, infected; d, days postinfection.

Fig 3

H5N1 and H1N1 influenza viruses differentially regulate the acute-phase response during infection. The diagrams show transcriptional changes for cytokine signaling pathway molecules in mouse lungs infected with either H5N1 virus (27) (A), 1918 virus (27) (B), or seasonal H1N1 virus (73) (C). Microarray data were background corrected and normalized by LOESS and quantile normalization. Expression values were represented as log₂ ratios of infected to respective mock-infected samples, where biological and technical replicates for a given condition were averaged. PathVisio was used for data visualization (143) of the acute-phase response signaling pathway from IPA (Ingenuity Systems). Official gene symbols are shown. Serum amyloid (Saa2 and Saa3) and serine (or cysteine) peptidase inhibitor (Serpina1d and Serpina3a) represent acute-phase response proteins. Red indicates that the gene expression is increased relative to that for the uninfected reference. Green indicates that the gene expression is lower than that for the uninfected reference. White indicates no change in gene expression, and gray indicates that a molecule was not detected by microarray analysis. inf, infected; d, days postinfection.

Fig 4

Mediators of the cytokine storm and the associated phenotypes with infection outcome. The diagram shows the functional roles of key cytokines and chemokines and their cognate receptors in the development of the main clinical outcomes associated with the cytokine storm. The redundancy of the cytokine and chemokine signaling pathways is emphasized.