Transcript Abstract - Centre for Genomic Research

Genetic Control of Cutaneous Leishmaniasis
1Noyes,
HA: 3Arana, FE; 3Rizzo, N; 1Kemp, SJ; 2Hutchinson, IV; 2Pravica V; 1Daly, D; 3Arana, BA;
1School of Biological Sciences, University of Liverpool, L69 7ZB, UK
2School of Biological Sciences, University of Manchester, M13 9PT, UK
3Medical Entomology Research and Training Unit, University del Valle, Guatemala.
Abstract
Sample Collection and Processing
Differences in expression of cytokines involved with the Th1 and Th2 responses are
associated with single nucleotide polymorphisms (SNP) in the genes that encode them..
We have genotyped SNP in a panel of genes for cytokines that are known to be important
in controlling the immune response.
Samples were collected from the following groups:
1) Patients with chronic cutaneous leishmaniasis of more than four months.
2) Subjects with a positive Montenegro skin test but no history of disease. These individuals
were expected to be resistant to the disease but not infection.
3) Individuals from the same endemic area with no history of disease and a negative skin
test. This is expected to be a heterogeneous group who may or may not have been
challenged
Significant differences were found in allele frequencies in the TNF-beta and IL4 genes.
Samples were collected in the Department of Peten in Guatemala. The Peten is an agricultural frontier
area where forest has been cleared for subsistence agriculture over the last forty years. Cutaneous leishmaniasis
(CL) due to Leishmania mexicana and L. braziliensis is endemic in this region.
Samples were collected from patients at the regular clinics run by the University del Valle (Guatemala) in
the town of Poptun in Peten. These clinics are announced on the local radio and provide a free service to all who
attend. Lesion scrapings for parasite identification and a 10ml blood sample are obtained from all patients who
have lesions consistent with cutaneous leishmaniasis. A Montenegro skin test (MST) is applied to all patients.
Parasites are identified by nested PCR (Noyes et al. 1998). Patients with a clinical diagnosis of CL were treated
with the WHO recommended dose of Glucantime.
Samples were collected controls by establishing clinics in communities which were known to be endemic
for CL. All members of the community were offered a test for anaemia and an MST skin test. 10ml blood
samples were taken from all volunteers over the age of 15. Samples were also collected from any cases of CL
that were discovered.
DNA was prepared from blood by ammonium chloride lysis of red cells and salt precipitation of DNA.
Cytokine alleles were identified in IL-6, IL-10, TNF-β, TNF-α, IFN-γ, IL4, IL15, TGF-beta by ARMS-PCR
(Perrey et al 1998). Additional SNP in GranzymeB and CCL5 were genotyped by SnapShot on and ABI Prism
3100.
A Chi squared test was used to identify significant differences between observed and expected allele
frequencies and genotype frequencies of individual loci.
SVD Clustering of arrays
Symptomatic; 18 hours
Asymptomatic; 18 hours
Symptomatic; 24 hours
Asymptomatic; 24 hours
Second Principal Component
Although microarrays are best known for gene expression analyses they are also a
powerful resource for identifying clusters of samples and have been widely used for
classifyin cancer cases. PBMC were taken from 30 MST+ve subjects with a history of
disease and 30 MST+ve subjects with no history of disease. PBMC were stimulated with
conA and harvested at 18 and 24 hours post stimulation. RNA was prepared from each
sample and combined into pools of five RNA samples. Each pool was labelled with cy3
and cy5 dyes to provide a dye flip replicate and hybridised to oligonucleotide microarrays
representing approximately 20,000 human genes (HGMP) using a common reference. The
normalised data was clustered using Single Value Decomposition in the MaxD
microarray data analysis package. After clustering the slides four groups could be
identifying corresponding to the cells from the two groups of patients that had been
stimulated for two lengths of time. The two groups of patients were separated on the first
principal component and the two times were separated on the second principal
component. Although the two groups of subjects overlapped a Student T-test indicated
that the difference between them was highly significant (p=0.00059). The subjects had
been selected at random irrespective of ethnic group, a Pearson correlation coefficient
indicated that there was no correlation between ethnic group and co-0rdinates on the first
principal component (r2 PCA1 = 0.027; (r2 PCA2 = 0.008).
IL4 allele frequencies
1
P<0.001
0.8
Diseased
0.6
Infected
0.4
Uninfected
0.2
0
Indigenous
Ladino
TNF beta
P<0.05
0.8
0.7
0.6
0.5
MST+ve w ith lesion
0.4
MST+ve no lesion
MST-ve no lesion
0.3
0.2
0.1
Significant differences were found
at the IL4-590 and TNF-b loci
between MST+ve subjects with no
history of disease and both subjects
with a history of active disease and
subjects with no history of disease.
In the case of IL4 this difference
was only observed in the “Ladino”
population of immigrant descent,
and there was no significant
difference in allele frequencies
among the different patient groups
of Indigenous descent. In the case
of TNF-beta the two ethnic groups
had opposite trends.
If confirmed these results would
indicate that these cytokines might
play a role in conversion on the
skin test but have no effect on the
development of disease.
0
Indigenous
Ladino
Discussion
Genetic risk factors for susceptibility to cutaneous leishmaniasis
First Principal Component
References and Acknowledgements
We gratefully acknowledge the participants in this study and Milo Henstermann and all the others who conducted
the field work. This work was funded by the Wellcome Trust.
Davies, C.R. and Gavgani, A.S.M. (1999) Age, acquired immunity and the risk of visceral
leishmaniasis: a
prospective study in Iran. Parasitology 119:247-257.
Noyes, H.A., Reyburn, H., Bailey, J.W., and Smith, D. (1998) A nested-PCR-based schizodeme method
for
identifying Leishmania kinetoplast minicircle classes directly from clinical samples and its
application
to the study of the epidemiology of Leishmania tropica in Pakistan. Journal of Clinical
Microbiology
36:2877-2881.
Perrey, C., Pravica, V., Sinnott, P.J., and Hutchinson, I.V. (1998) Genotyping for polymorphisms in
interferongamma, interleukin-10, transforming growth factor-beta 1 and tumour necrosis factor- alpha genes: a technical
report. Transplant Immunology 6:193-197
Turner, D.M., Williams, D.M., Sankaran, D., Lazarus, M., Sinnott, P.J., and Hutchinson, I.V. (1997) An investigation
of polymorphism in the interleukin-10 gene promoter. European Journal of
Immunogenetics 24:1-8.
There is no evidence of any difference in susceptibility to leishmaniasis between ethnic
groups in Guatemala, with a prevalence of history of leishmaniasis about 10% in both the
indigenous and immigrant populations. However there was some evidence that the two
groups might have different risk factors for conversion on the skin test. Most notably at
the IL4-590 locus where there was a significant risk of conversion on the skin test
associated with the C allele in the indigenous group but not the Ladino group. It is not
uncommon for associations found in one ethnic group to be missing in another group.
However , the sample sizes in this study were small and it will be important to determine if
the results observed here are reproduced in other samples form the same population as
well as in other ethnic groups to determine whether these observations were a real risk
factor or whether they are artefacts of small sample sizes.
Genetic basis for conversion on the skin test?
Since the micoarray study was undertaken on cells stimulated with ConA rather than a
Leishmania antigen the results provide evidence that there is a genetic component to the
risk of conversion on the skin test. This may have important implications for the
interpretation of the results from this widely used diagnostic and epidemiological test.