Transcript Host genetics and disease resistance

Genomics approaches to trypanosomiasis resistance
Breeds of cattle which are resistant and susceptible to pathology following Trypanosoma congolense infection, show
differences in gene expression following challenge. Examination of the networks of genes responding most
differently between these genotypes provides insights into the biology of the response to infection.
Agaba M1
Anderson S4
Archibald A4
Brass A2
Where differentially used networks include genes within QTL for resistance, those genes become functional
candidates irrespective of whether they are differentially expressed themselves. The diversity of cattle and disease
challenge across Africa provides a unique natural experiment which can be used to evaluate these genes for
evidence of a functional role in trypanosome resistance.
The responding networks provide new ways of examining the pathology, raising hypotheses and suggesting novel
interventions which may be unrelated to the present diversity represented by the QTL.
N’Dama (tolerant)
Boran (susceptible)
Gibson J5
Hall L4
Hanotte
Hulme
O1
H2
Kemp SJ1,3
Mwakaya
J1
Noyes, HA3
Ogugo M1
15
10
5
0
-5
-10
-15
QTL with p<0.0043 FDR(10%)
PCV
Body weight
Parasitaemia
15
10
5
0
-5
-10
-15
A cross between relatively
susceptible Boran and resistant
N’dama has shown that each
breed carries 5 different QTL for
resistance, and these have been
mapped in an F2 cross.
Hanotte et al PNAS 2003 7443-7448
Twenty N’Dama (tolerant) cattle and 20 Boran (susceptible) cattle were challenged with a lethal dose of T. congolense. Liver
biopsy samples were taken from each individual in specified days prior to and post infection such that at each time point there
were samples from at least 5 Boran and 5 N’Dama.
The mRNA profiles were assayed using the Affymetrix 24K probe sets (B). The gene data were fed into an analysis workflow (C)
that integrates the expression measures, gene ontology, QTL information, and gene pathways data.
Rennie C3
Addresses
1International Livestock
Research Institute,
Box 30709 - 00100, Nairobi
Kenya
2The
University of
Manchester
LF8 Kilburn Bldg, Oxford Rd
Manchester M13 9PL
UK
3School
of Biological
Sciences,
University of Liverpool,
Liverpool, L69 7ZB, UK
4Roslin
Institute,
Roslin,
Midlothian, EH25 9PS, UK
5The
Institute for Genetics &
Bioinformatics, Hawkins
Homestead, University of
New England, Armidale,
NSW 2351, Australia
Acknowledgements:
We thank all the staff at the
ILRI large animal facility and
all colleagues in the
Welcome Trust Consortium
This work was supported by
the Wellcome Trust.
BIG DIFFERENCES BETWEEN GENOTYPES AND OVER TIME. Between 600 and 750 probes were differently expressed
between infected and uninfected cattle. Principle component analysis of the expression data clearly shows genome-wide
differences between the transcriptomes of tolerant () and susceptible ( ) cattle (Top Right, PCA component 3) and some of these
differences are associated with the presence and progression of trypanosome infection (Top left, PCA component 1).
GeneGo was used to identify networks amongst the
genes that were differentially expressed. The figure
(right) shows the largest network of connected genes that
were differently expressed by trypanosome infected
resistant (N’Dama) and susceptible (Boran) cattle.
STAT3 and c-Fos have the most connectivity. STAT3 is a
transcription factor which is activated in response to the
IL-6 family of cytokines and is involved in the acute
phase response in the liver
Interestingly, STAT3 is modulated by RAC1 which is in
turn controlled by VAV1 and ARHGAP15 which are both
located in the QTLs controlling trypanotolerance.
 Different cattle genotypes show differences in gene expression before and during trypanosome infection.
 Trypanosome infection induces profound changes in the steady state level of many genes.
 Network-based analysis reveals some highly concordantly responding networks
 Genes that are within QTL and also within highly connected networks are strong candidates for QTL genes
 Some differentially expressed genes are highly connected to many other differentially expressed genes. These are key points
for intervention.