Coxiella burnetii NMII infections were initiated as described and fixed at 0, 8, 16, 24, 48, 96, and 168 hpi with 4% paraformaldehyde, 0.05% Tween-20 in phosphate-buffered saline for 15 min at room temperature. Indirect immunofluorescent antibody (IFA) analysis was performed by dual staining as described previously (Morgan et al., 2010). Micrograph images were captured via a Nikon DS FI1 camera on a Nikon Eclipse TE 2000-S microscope at × 400 magnification, with nis-elements f 3.00 software. A magnification of × 400 was used as opposed to × 600, as used previously (Morgan et al., 2010). Micrograph capture settings were uniform for all images. Using a modification of a method used previously in C.

burnetii studies that uses relative pixel ratios in sample quantitation (Zamboni et al., 2001), each micrograph image was analyzed using imagej version EPZ015666 manufacturer 1.42n (Wayne Rasband, NIH) software. Five fields of view from each time sampled (three biological samples of each) were digitally captured. The matching Alexa Fluor® 555 and Alexa Fluor® 488 images were stacked (paired) and converted to gray scale (8 bit). No fewer than five regions of interest (ROI) were blindly selected from each field of view of the 555 nm grayscale images. This provided at least 75 ROIs for each time point analyzed. Saturated regions of an image were not selected and ROI size varied

BIBF 1120 molecular weight depending on the PV size. The pixel densities within the identical ROIs from each stacked image were then measured

as published previously (Collins, 2007). The mean pixel densities were then compared to obtain the 488 : 555 ratio for each ROI. These individual ratios were then averaged (≥75 individual ratios/time point) to determine the relative expression of IcmT to whole C. burnetii NMII. The final 488 : 555 (IcmT : C. Ureohydrolase burnetii) ratio for each time point was then divided into the 0 hpi ratio to obtain the final IcmT relative expression levels. The statistical significance between each time point was evaluated using single-factor anova with a 95% confidence interval using ms excel 2007 (Microsoft). The C. burnetii T4BSS RI gene linkage map suggests that three groups of transcriptionally linked genes exist (see Fig. 1a). These include: (1) icmXCBU1651icmW, (2) icmVdotACBU1647, and (3) CBU1646dotBdotCdotDicmSicmT. To demonstrate transcriptional linkage between the genes within each group, RT-PCR analysis was performed using oligonucleotide primers (see Table 1) designed to span intergenic sequences and/or adjoining ORFs. The diamond-ended lines in Fig. 1a indicate the position of primers and DNA products that would result from RT-PCR amplification. Using total RNA harvested from Vero cells infected with C. burnetii NMII as a template, amplification products were observed (Fig. 1b) for each linkage region: (1) icmW–icmX, (2) icmV–dotA, dotA–CBU1647, and (3) icmT–dotD, dotD–dotB, dotB–CBU1646 (Fig. 1b).