Global gene expression profile of chicken cecal tonsil in response to campylobacter jejuni challenge in broiler lines

It is essential to understand host response to Campylobacter jejuni infection in order to genetically improve resistance to its colonization in chickens. A custom Agilent chicken 44K array was used to examine gene expression profiles after Campylobacter jejuni infection of two broiler lines (A and B). Day-old chicks were orally inoculated with C. jejuni. After day 7 post-infection, the cecal tonsil was collected for total RNA isolation and cecal content for bacteria burden quantification. Twenty highest and lowest bacterial burden birds and non-infected birds within each line were used to pool four biological replicates for each group. The pair comparisons among high, low bacterial burden, and non-infected group were used. The signal intensity of each gene was normalized by LOWESS method. A mixed model including the fixed effects of dye, line, treatment and line × treatment interaction, and random effects of slide and array was used to identify differentially expressed genes at P < 0.001 by SAS program. Within line A, there were 61, 163, and 90 genes significantly differentially expressed between high and low bacterial burden, high bacterial burden and non-infected group, and low bacterial burden and non-infected group, respectively; 2637, 1684, 561 genes within line B, respectively. The results suggested that genetics, treatment and genetics × treatment interaction played important role in gene regulation of C. jejuni infection. The findings in the current study will lead the identification of potential candidate genes for genetic resistance to C. jejuni infection in chickens. Keywords: diease state analysis Chickens in two broiler lines were inoculated with 10^5 cfu C. jejuni on one day after hatch. The cecal content and cecal tonsil was collected and bacterial number in cecal content was counted on day 7 after inoculation. Twenty samples were seperated into 3 groups (high burden, low burden, and control) based on bacterial burden of cecal content in each line, 5 samples were mixed randomly into one pool. A dual color, balanced design was carried on for all samples. There comparisons were used in each line, C/H, H/L, L/C, totally, four biological replicates in each line. A Dye swap was used in each pair of comparisons including AC/AH, AH/AL, AL/AC; BC/BH, BH/BL, and BL/BC. Background subtracted signal intensity were collected from 24 arrays and normalized for data analysis.