Prediction and verification of NF-κB targets in the porcine MHC through the use of sequence similarity and pathway inhibition
With the advent of high throughput technologies for both the sequencing of genomic DNA and the measure of the expression of RNA a tremendous amount of information has been generated and deposited into public databases. This large amount of data has led to the better understanding of how a genome is organized, the number of regions encoding information for transcripts, as well as how the amount of these transcripts change due to various perturbations a cell or organism encounters, whether it be an outside stimuli, such as bacteria or viruses, or internal, such as a mutation within the genome.
Some species, such as the human and mouse, have had a significant amount of sequencing completed, leading to excellent reference genome sequences, as well as these sequences being well understood at the function and structure level, termed gene annotation. However, for most vertebrate species, their genomes are in various states of completion; from being nearly completed with partial annotation, like the pig, to having only portions of their genomes completed, such as Alatina moseri, a species of Hawaiian jellyfish. For these species the amount of direct annotation is greatly lacking compared to that of other species, such as human and mouse. When annotation is lacking for one species, it is possible to leverage the information already obtained for closely related but better-studied species by comparing sequences across species and identifying similar regions between them, allowing the annotations of these regions to be inferred across species.
Once a species has sufficient sequence annotation, high throughput expression data, such as that from microarrays, can be better understood. One area of research that is under development, which can utilize high throughput expression measures, is understanding how a set of transcripts changing together in response to perturbations in the environment is controlled by specific proteins, called transcription factors, such as NF-κB. NF-κB is an important transcription factor, having a role in a variety of cellular functions, such as mounting a response to infection and preventing cell death by inhibiting apoptosis. While some transcription factors, like NF-κB, have been well studied and many of its target genes identified, this identification is typically done one or a few genes at a time. However, as more genomes are sequenced, better algorithms developed for identification of possible targets, and new biological techniques optimized, the ability to predict and verify targets is also moving toward high throughput. In order to create more reliable gene annotation for the pig, raw porcine sequences were assembled into more full length sequences to create an accurate base for which to compare to other species, as well as identify possible sequence variation within the assembled sequences. This annotation was then used in a high throughput experiment to look for genes changing expression due to an inoculation of Salmonella choleraesuis in pigs, and to determine which genes are potential NF-κB targets. Then, potential target genes found in an immune related region of the genome were tested in response of bacterial endotoxin either in the presence or absence of an NF-κB inhibitor. The ability of NF-κB to bind to their promoters was also tested using a labeled EMSA probe. Using these two methods, we show the murine H2-Eb1 and Trim26 and porcine C2 and UBD are novel targets of NF-κB and that such bioinformatic predictions can be confirmed using molecular assays.