A Common Genetic Fingerprint in Leprosy and Crohn’s Disease?
Alan Kennedy
Erwin Schurr, Ph.D. and Philippe Gros, Ph.D.
New England Journal of Medicine: Editorial: December 16, 2009
http://content.nejm.org/cgi/content/full/NEJMe0910690
The cause-and-effect relationship between severe infections and death
suggests that microbial pathogens are evolutionary sculptors of the
genome. However, the genetic component of susceptibility to infections
in the general population is complex and heterogeneous and is
modulated by environmental factors such as determinants of microbial
virulence. Thus, it is a challenge to identify specific genetic
effects in human populations. Availability of the human genome
sequence, combined with knowledge of genetic variation, has
facilitated the genomewide association study, a powerful approach to
detecting genetic associations. In this issue of the Journal, Zhang
and colleagues (1) describe a genomewide association study of leprosy,
a bacterial disease.
Leprosy manifests with a broad pathologic spectrum. At one end is the
localized paucibacillary form, characterized by a small number of
hypopigmented, anesthetic skin lesions; at the other end is the
disseminated multibacillary form, involving numerous skin lesions with
a high bacillary load. Paucibacillary infection is associated with
immune responses mediated by type 1 helper T (Th1) cells (involving
the production of interferon-gamma and interleukin-2) that promote
granuloma formation and limit bacterial replication and dissemination.
The multibacillary form, in contrast, is associated with the Th1
polarization of the immune response (and the production of
interleukin-4 and interleukin-10), which promotes uncontrolled
bacterial replication and more severe pathology.
Population studies and studies of twins have established that there is
a genetic component to susceptibility to leprosy.(2,3) Linkage and
association studies have implicated variants of the HLA-DR region,
PARK2 (encoding parkin), LTA (encoding lymphotoxin alpha), and
chromosome 10p13 in conferring susceptibility to leprosy in
independent populations.(3)
Zhang and colleagues describe the results of genomewide scanning in
persons with paucibacillary or multibacillary forms of leprosy from
eastern or southern China. The authors compared the prevalence of each
genetic marker - in this case, each single-nucleotide polymorphism
(SNP), which is currently the marker typically used in genomewide
association studies - in 706 case patients and 1225 unaffected
persons. A total of 93 SNPs were shown to have a significant
association with leprosy. These SNPs were then tested in three
replication sets of more than 3000 patients and nearly 6000 controls
from eastern or southern China. The data implicate CCDC12 (the gene
encoding coiled-coil domain containing 122), C13orf31 (encoding
chromosome 13 open reading frame 31), NOD2 (encoding nucleotide-
binding oligomerization domain containing 2), TNFSF15 (encoding
tumor necrosis factor [ligand] superfamily member 15), RIP2K
(encoding receptor-interacting serine-threonine kinase 2), and the
HLA-DR-DQ locus. Several of the proteins encoded by these genes are
involved in microbial sensing and in the early innate immune and
inflammatory responses. NOD2 recognizes a component of the
mycobacterial wall, and stimulation of NOD2 results in the recruitment
of RIPK2 and indirectly prompts the activation of the transcriptional
regulator nuclear factor Kappa-B (NF-kappaB4) - which in turn
activates the transcription of genes encoding proinflammatory
cytokines including TNFSF15. On the surface of phagocytes, HLA-DR
molecules present bacterial antigens to CD4+ T cells to initiate Th1-
cell polarization.
Human genetic studies have implicated both IL12B (the gene encoding
interleukin-12beta) and NOD2 in increased susceptibility to
mycobacterial disease, and mouse mutants lacking either Nod2, Ripk2,
or Infg (encoding interferon-gamma) are highly susceptible to
tuberculosis.(5,6,7) The fact that such genes are now implicated
through a genomewide association study of leprosy not only validates
this approach to studying the disease, but also raises interest in the
newly implicated genes (e.g., CCDC12 and C13orf31 ), the functions
of which are not known.
Another interesting aspect of the study is that variation in some of
the implicated genes is known to be associated with bowel inflammatory
conditions. A frame-shift mutation in NOD2 has been identified as a
strong susceptibility factor for Crohn's disease; additional NOD2
mutations have been discovered not only in persons with Crohn's
disease but also in those with Blau's syndrome and in those with early-
onset sarcoidosis.(8) Likewise, variants of TNFSF15 and IL12B have
been associated with Crohn's disease.(9) These findings are consistent
with studies of mouse models that have also established a role for
Nod2, Ripk2, and Nfkb in intestinal homeostasis and colitis.(10,11)
Together, these studies establish a strong genetic and functional link
between susceptibility to leprosy and predisposition to Crohn's
disease.
Although the results described by Zhang and colleagues are exciting,
additional experiments are required to validate and refine their
conclusions. Genomewide association studies are fairly crude. For
example, a regulatory SNP in the LTA gene, previously identified as
a leprosy risk factor,(12) was not tested in the platform used by
Zhang and colleagues. Moreover, linkage disequilibrium (a state in
which genetic markers - typically those in close physical proximity -
are more likely than not to be inherited together) can extend over
large intervals, with the main genetic effect located within a region
that is a considerable distance from the SNPs showing the disease
association. As known SNPs increase in number, the dissection of local
effects of linkage disequilibrium will become more accurate, and
testing for the presence of associations in groups of persons of
differing ancestries will shed light on the extent to which these
findings of Zhang and colleagues apply to other populations.
How do we move from P values to understanding pathogenesis and
response to infection and, finally, to clinical outcomes? Although
genomewide association studies rarely identify causative genetic
lesions, they do point to specific genes and biologic pathways that
can be targeted for pharmacologic intervention, irrespective of the
mechanism underlying genetic susceptibility. A particularly attractive
aspect of the study by Zhang and colleagues is the apparently narrow
focus of the genetic control, which highlights early antigen sensing
and signaling in the pathogenesis of both leprosy and Crohn's disease.
It is tempting to speculate that these common genetic signatures
support, albeit indirectly, the proposal that a proportion of Crohn's
disease cases may have a mycobacterial cause.(13,14) Irrespective of
its strength, such a link may broaden the therapeutic treatment
options for both diseases.
In comparison with the study by Zhang and colleagues, genomewide
association studies of susceptibility to malaria(15) and to infection
with the human immunodeficiency virus(16) suggest that the
contribution of common genetic variants is more limited. Why would
this be so? The genomic variability of Mycobacterium leprae isolates
is very small, and M. leprae has undergone substantial reductive
evolution, possibly through adaptation to its human host. This may
suggest that larger host genetic effects in infectious disease reflect
decreased pathogen variability. In this view, pathogens with greater
genetic variability, such as M. tuberculosis, will give rise to a
more complex genetic architecture of host susceptibility.
----------------------
I've started to publish these updates on a blog: http://crohn.ie
Alan.