We characterized the differential effects of 6 cytokines on gene expression in a mixed population of human immune cells; defining for the first time the conserved transcriptional response to IFNω, together with IFNα and β, and revealing major similarities in IFNγ and IL12 transcriptional programs. We and many others have examined gene expression profiles in peripheral blood in vivo
and ex vivo
in order to identify both gene expression signatures and specific regulatory pathways activated in health-associated homeostasis and in disease. Data derived from carefully controlled experiments establishing the transcriptional response to specific stimuli in relevant cell populations have become an important part of the framework used to interpret patterns of gene expression observed in these complex studies. The response to interferons has gained a prominent central role in the study of many immune processes. We have observed an interferon-response gene expression signature associated with shock in dengue infection 
, with early pre-clinical responses to Ebola 
, with the response of leukocytes to Neisseria meningitidis 
, and with a pattern of gene expression that discriminates Kawasaki Disease and acute adenovirus infection 
. The detailed description of the shared and unique features of the transcriptional programs of the cytokines used in this study, and of the contribution of individual cell types to the gene expression patterns observed in a mixed cell population constitutes a valuable resource that has facilitated our analysis of these previously-published data sets, and will enable researchers to better delineate regulatory networks, elucidate pathways that are disrupted during disease, and deconvolute gene expression patterns in cell populations exposed to multiple complex stimuli.
Other groups have examined the response to interferons in non-immune cell populations. Many of the observations of Der et al
who investigated the expression profiles of HT1080 cells (a human fibrosarcoma cell line) after 6 h stimulation with interferons IFNα, β, and γ by microarray analysis, were confirmed in this study. A core set of established IFN-regulated genes were identified by both analyses; however, there are subsets of genes that appear to be induced differentially in these two studies. This may be due to technical differences such as concentration of interferons, microarray platform and gene annotation, but may also reflect cell-specific (professional immune vs non-immune) responses to IFN stimulation 
. It is also possible that differences reflect donor-to-donor variation in interferon-mediated responses. However, the overall similarity in findings, in addition to the results from studies investigating inter-individual variation in gene expression patterns, indicate that the stimulus rather than a donor-specific background is likely to be the primary feature driving transcriptional responses of peripheral blood immune cell populations 
. Similarly, comparison of our results with those from studies of IFNγ stimulation of primary endothelial cells and HT1080 cells 
revealed broad concordance, with a few exceptions. A recent study by Indraccolo et al
identified 41 genes to be more highly induced by IFNα compared to IFNγ in human umbilical vein endothelial cells; as in our study, the majority of these genes were predicted to be involved in antiviral responses. Six genes were induced to a greater degree by IFNγ than by IFNα; of these, CXCL9 (MIG) and ubiquitin D (UBD or FAT10), both of which are involved in targeting proteins for proteasomal degradation 
, were also identified in the present study. Therefore, the chemokine, CXCL9, and UBD, which is a marker of immune activation in hepatocellular carcinoma and colon carcinoma 
, are preferentially induced by type II interferon in both professional immune and non-immune cells.
A parallel analysis of IFNγ responses of isolated subsets of the cells comprising the PBMC pool allowed us to begin dissecting the contribution of specific cells in this complex tissue. The transcriptional response to IFNγ in T cell, B cell and NK cellular populations was limited in comparison to the large number of genes induced in purified monocytes, where RHOH and CISH are predicted to have novel regulatory roles in controlling monocyte activation. These transcript abundance datasets may be utilized in combination to examine which particular cell types are responding to specific stimuli in dynamic scenarios. For example, the transcription factors C/EBPα, involved in myeloid cell differentiation 
and associated with acute myeloid leukemias 
, and MHC2TA (CIITA), the major regulator controlling expression of MHC class II molecules 
, were both preferentially induced by type II IFN (in PBMCs compared to type I IFNs and TNFα). These genes were also up-regulated with time in the monocyte population alone (in response to IFNγ treatment), suggesting that the expression of these major mediators of cell activation in a complex model of infection may be due primarily to the response of monocytes to IFNγ. This study also highlighted the plasticity of the cellular transcriptional response to stimulation, with the induction of ostensibly cell type-specific genes in alternative cell types after treatment with IFNγ. Comparison of the expression patterns derived from mixed and purified immune cell populations after IFNγ treatment illuminated differences in the way in which cells responded to a stimulus in isolation versus a complex environment, with a number of genes only induced by IFNγ stimulation in complex cellular populations. The immune response to pathogens requires the coordinated interaction of multiple cells, and the presence of different cell populations; thus, the cross-talk between populations may be more representative of in vivo
functioning than are the findings from isolated cell populations. Further investigation into these interactions may provide novel insights into cellular activation and complex responses measured from mixed populations.
The focus of this study has been on the defined actions of cytokines in the activation of the immune system. While infection is one setting in which these actions may be relevant, these data may also help to elucidate additional functional roles of interferons, such as in tumor suppression. Exploration of the transcriptional programs initiated by interferons in heterogenous and homogenous cell populations, extends our understanding of the global actions of these major mediators of inflammation, improving our comprehension of disease states, immune cell signaling cascades, the immunomodulatory mechanisms of infectious agents and potential recombinant therapies.