Enrichment analysis of transcription factors binding sites in blood cells in response to ingestion of bovine milk by healthy adult humans

by Sagaya mary francina


organisms respond to diets or dietary stimuli by
regulating gene expression. Transcription factors play a central role
in regulating the expression of the genes that
can modulate the physiological
processes maintaining health. Milk
provides unique nutritional and health benefits to mammals, including
humans. As the analytical tools become available there is a growing
interest in investigating the nutritional properties of milk at the
molecular level to identify the underlying regulatory mechanisms. In
that context, we have analysed the promoter regions of the genes that
were found to be significantly induced in response to milk and yogurt
ingestion by human subjects. In particular we have searched for
binding motifs for transcription factors involved in the regulation
of these genes.


and methods

transcription factor binding site (TFBSs) enrichment analysis was
carried out using pre-existing data from a transcriptomic analysis
already performed in our laboratory. Briefly, a randomized controlled
single blinded crossover study was conducted involving six healthy
male individuals. The duration of the study was for 8 days involving
a three days of run-in period followed by two days of intervention
(on days 4 and 8) and three days of washout period. During the two
intervention days, the subjects were given either milk or yogurt
after an overnight fasting. Blood samples were collected before (0h)
and after dairy ingestion (milk and yogurt) at 2, 4, and 6 hours on
day 4 and on day 8. The blood cell transcriptome was analyzed using
whole human genome arrays from Agilent.

A linear contrast kinetic analysis identified 575 and
625 (FDR 12.5%; P ≤
0.002) statistically significant and
differentially expressed transcripts in the milk and yogurt groups,
respectively. The differentially expressed genes were then clustered
into up-regulated and down-regulated genes based on their gene
expression pattern. Additionally, Gene ontology (GO) based functional
groups common to milk and yogurt were also identified. The promoter
sequences (upstream +1000bp to downstream -200bp) of up-regulated,
down-regulated genes as well as functional groups (translation,
mitochondrial part, immune response, inflammation and apoptosis) were
retrieved from ENSEMBL using Toucan version 3.1software. CLOVER
software was used to screen the JASPAR CORE (2005) database, which
contains 123 TFBS motifs. We set the
parameters of CLOVER as 2,000 randomizations and a p-value
threshold of 0.05. For the estimation of p-values,
we supplied 3 sets of background sequences to the algorithm, which
estimates p-values for
each background set separately. The background sequences included
were chromosome 20 sequences (7 sequences, 59424990 bp, 44.1% C+G),
CPG sequences (27555 sequences, 20891518 bp, 68.8% C+G) and 2000bp
upstream of human promoter sequences (17516 sequences, 35032000 bp,
49.8% C+G). Motifs were considered significantly enriched only when
p-value ≤
0.05 against all the three background.



bioinformatics analyses showed that, out of 123 non-redundant JASPAR
motifs, 10-15% are enriched in datasets composed of postprandial
genes responding to milk and yogurt. These motifs are recognized by
transcription factors belonging to the classes HMG, zinc-finger C2H2
(broad-complex and hunchback), MADS, forkhead, ETS and AP2. Most of
the motifs were enriched in both the up- and down-regulated sets of
genes (Figure 1). The ingestion of yogurt also resulted in the
enrichment of motifs similar to the ones observed upon the ingestion
of milk. Taken together these results suggest that the postprandial
response of the human organism to the diet, in particular to dairy
products, is similar at the level of transcriptional regulation.
Specificity was nonetheless also observed for a subset of binding
motifs. In particular, down-regulated functional groups involved in
apoptosis, immune response and inflammation showed binding motifs
enriched in the transcription factors NF-kB, RelA, AP-1, IRF-1 and
members of the forkhead family (FoxL1, Foxq1). Interestingly, the
motif for the transcription factor GABPA (NRF-2), the master
regulator of the antioxidant response, is selectively enriched in the
set of up-regulated genes. This motif is also enriched in the
functional groups containing the up-regulated genes involved in
translation and in the group of genes associated with a mitochondrial
subcellular location. All the motifs identified using CLOVER, along
with their motif scores and p-values
is provided as supplementary material 1.



One of
the challenges of computational biology is to identify genomic
binding sites for transcription factors. The identification of these
binding sites allows the elaboration of more accurate gene networks
and, consequently, an understanding of the regulation of complex
biological pathways. One of these biological pathways is the
postprandial metabolic response of the organism to dietary
components. As a starting point towards the elaboration of gene
networks involved in this process, we identified regulatory factors
such as GABPA, forkhead, HMG, broad-complex which may be involved in
the postprandial transcription regulation in response to dairy
ingestion. In the context of the diets selected in our study, the
significant enrichment of the motif binding to the transcription
factor GABPA leads to the identification of a key molecular target
that may mediate the anti-oxidative properties of milk, a biological
activity that is also closely linked to its reported
anti-inflammatory effects. Overall, our results provide a holistic
picture of key regulatory factors potentially involved in the
postprandial anti-oxidants/anti-inflammatory properties of dairy



important finding in this paper shows enriched GABPA mediated gene
expression in up-regulating genes. Thus,
activation of GABPA in immune cells in response to milk ingestion
might be a useful therapeutic approach in conditions where there is
an elevated level of oxidative stress and associated inflammation.


Figure 1