Early indicators of exposure to biological threat agents using host gene profiles in peripheral blood mononuclear cells

Abstract

Background: Effective prophylaxis and treatment for infections caused by biological threat agents (BTA) rely upon early diagnosis and rapid initiation of therapy. Most methods for identifying pathogens in body fluids and tissues require that the pathogen proliferate to detectable and dangerous levels, thereby delaying diagnosis and treatment, especially during the prelatent stages when symptoms for most BTA are indistinguishable flu-like signs.

Methods: To detect exposures to the various pathogens more rapidly, especially during these early stages, we evaluated a suite of host responses to biological threat agents using global gene expression profiling on complementary DNA arrays.

Results: We found that certain gene expression patterns were unique to each pathogen and that other gene changes occurred in response to multiple agents, perhaps relating to the eventual course of illness. Nonhuman primates were exposed to some pathogens and the in vitro and in vivo findings were compared. We found major gene expression changes at the earliest times tested post exposure to aerosolized B. anthracis spores and 30 min post exposure to a bacterial toxin.

Conclusion: Host gene expression patterns have the potential to serve as diagnostic markers or predict the course of impending illness and may lead to new stage-appropriate therapeutic strategies to ameliorate the devastating effects of exposure to biothreat agents.

Figure 1

(a) B. anthracis exposure to PBMC from 3 different donors. Data shown are from exposures at 2, 4, 8 and 24 h. Data from each exposure time period were separately evaluated in order to identify common trends among the three donors (males, ages 61, 41, and 27 years old (respectively) with diverse ethnicity). (b). Comparisons of gene profiles for 8 pathogenic agents. Human PBMC were exposed to each of these pathogenic agents for at least 3 appropriate time periods. RNA was isolated, and the reverse transcript hybridized to cDNA arrays. Unique gene patterns were induced by BTAs. Cluster diagram of gene expression patterns use Gene Spring analysis to illustrate groups of genes that show discriminatory patterns for various threat agents. These genes were compared for their expression patterns across all agents and time points. Red is up regulated, blue is down regulated and black is no change compared to the control sample. The expression patterns illustrate how one can differentiate pathogenic agents by selection of sets of gene expression patterns for examination. (Gene accession ID numbers, rather than gene names, are all provided legibly in the graphs of Additional files).

Figure 2

PCA relational analysis to show how the gene profiles (various exposure times) cluster for each toxin (a) and the relationship among the various pathogens (b). Human PBMC were exposed to each of these pathogenic agents for at least 3 appropriate time periods. RNA was isolated, and the reverse transcript hybridized to cDNA arrays.

Figure 3

Functional categories of genes that show similarities and differences between these pathogens. Accession numbers are shown associated with Figure 2, Additional files. Human PBMC were exposed to each of these pathogenic agents for at least 3 appropriate time periods. RNA was isolated, and the reverse transcript hybridized to cDNA arrays.

Figure 4

Comparison of host gene responses in vivo and in vitro exposures to anthrax. Gene expression profiles in PBMC from healthy human donors exposed to anthrax spores in vitro were compared with gene expression patterns obtained in PBMC taken at 24, 48 and 72 hr after exposure of NHP to anthrax spores by aerosol challenge. (a) Gene cluster analysis of significantly altered genes in vivo. (b) comparison of gene expression profile between in vivo and in vitro exposures.

Figure 5

Clustered sets of genes to illustrate stage-specific vs. commonly expressed genes for in vivo exposures of NHP at 24, 72 h or at both time periods.

Figure 6

Confirmation of selected gene changes by RT-PCR with in vitro and in vivo samples for IL-6 (a) and Transducin beta-1 subunit, GNB1, (b).

Figure 7

Expression of GBP-2 and IL-6 genes after in vivo SEB exposure of NHP for 30 min. Gene expression profiles in PBMC from NHP exposed to SEB for 30 min. RNA samples were isolated and used in the PCR assays using primers specific for IL-6 and GBP-2.

Figure 8

Ordered genes and resulting percentage correct classifications. For a given number of genes, a set of genes most able to discriminate between disease states is selected. Simulations of noisy readings of patient gene expression levels are performed for varying levels of noise. Each colored line plots the percentage of correct classifications versus the number of genes used to make the classification for one particular percentage of random values in the simulated readings. With no noise, very few genes are required to discriminate perfectly. With high noise levels (say, 50%), even 1000 genes cannot reliably discriminate well.

Figure 9

Expression patterns of genes that were at baseline levels in all controls and showed unique expression patterns in (a) BoNT-A exposures or (b) B. melitensis exposures compared to all 8 pathogens.

Figure 10

Real-time PCR determination of gene expression in response to each of 8 pathogenic agents. Primers were designed for these 3 genes and 18S, which was used as a reference gene for comparison of these 3 test genes. GBP-2 was a gene that was identified as being massively up regulated by SEB using differential display PCR (Mendis, et al) and was of particular interest to us.