 |
Herpes Simplex Virus Encephalitis: A Novel Group of Primary Immunodeficiencies
Summary: Jean-Laurent Casanova is conducting research on the potentially fatal condition known as herpes simplex encephalitis, which, for an unknown reason, develops from herpes simplex virus-1 in a small percentage of infected children. His search for a candidate gene and an understanding of the underlying immunodeficiency of this disease in families may have important medical and biological implications.
Herpes simplex virus-1 (HSV-1) is a double-stranded (ds) DNA virus that infects more than 90 percent of young adults worldwide. It enters the body via the oral or nasal epithelia, where it infects neurons and establishes a latent infection in sensory ganglia. HSV-1 causes asymptomatic infection or benign, self-healing disease in most individuals, with gingivitis and stomatitis during primary infection and herpes labialis during reactivation. In rare cases, HSV-1 causes life-threatening herpes simplex encephalitis (HSE). In children, HSE typically occurs during primary infection, whereas in adults it often results from reactivation. The virus reaches the central nervous system (CNS) via the trigeminal and/or olfactory nerves, with no documented viremia or viral lesions elsewhere, either in epithelia (oral epithelium) or internal organs (liver). Until the introduction of acyclovir in the early 1980s, HSE was almost invariably lethal in children. Even now, HSE is a devastating illness, and most survivors still present profound neurological sequelae, such as mental retardation and recurrent seizures.
The pathogenesis of this severe disease remains unclear. It is unknown why only a small minority of children infected with HSV-1 develop HSE (about 1/200,000 individuals per year). It is also intriguing that the lesions observed are restricted to the CNS. Moreover, these children are otherwise healthy and are not prone to other unusually severe infectious diseases, including those of viral origin. Finally, it is remarkable that children with conventional primary immunodeficiencies, including conditions characterized by a lack of most myeloid and lymphoid leukocyte subsets, such as agammaglobulinemia (lack of B cells), congenital neutropenia (lack of granulocytes), severe combined immunodeficiency (lack of T cells), and reticular dysgenesia (lack of leukocytes), are not prone to HSE. Although children with severe primary immunodeficiencies are vulnerable to a wide range of infections, they have never been reported to contract HSE. Thus, patients with HSE display selective vulnerability to HSV-1 replication in the CNS.
We hypothesized that, in at least some children, HSE may result from an unidentified form of primary immunodeficiency associated with a Mendelian susceptibility to HSV-1. This hypothesis arose from our research in general and from our more global theory that life-threatening infectious diseases before puberty result from monogenic traits (as opposed to polygenic inheritance). The high frequency of consanguinity among the parents of patients (10 percent) in our recent retrospective French genetic epidemiological survey of pediatric HSE is strongly consistent with this hypothesis. Given that children with HSE mount a detectable HSV-1–specific T cell and B cell response, we hypothesized that the underlying inherited defect in these children, if there is one, is likely to affect innate rather than adaptive immunity. We first used a candidate gene approach to investigate an important element of antiviral innate immunity: type I and type III interferons. More recently, we used a genome-wide screening approach for homozygosity mapping of disease-causing genes in consanguineous families.
The candidate gene approach rapidly yielded results, with the serendipitous identification of the molecular genetic basis of HSE in two children with an exceedingly rare presentation of disseminated mycobacterial disease and HSE. We had previously shown that genetic defects affecting the production of, or response to, type II interferon (IFN-γ) were associated with a selective predisposition to severe mycobacterial disease. One of our pediatric patients with mycobacterial disease died of HSE. We explored the IFN-γ signaling pathway in this patient, who was found to have complete STAT-1 deficiency. The lack of response to IFN-γ accounted for the mycobacterial disease. Unlike the cells of patients with a specific defect of IFN-γ–mediated immunity and mycobacterial disease without HSE, the cells of this patient did not respond to type I interferons, probably accounting for the development of HSE. Shortly thereafter, we identified a second child with mycobacterial disease and HSE. He presented a mutation in NEMO that impaired the production of both type I and type II interferons. The data obtained for these two patients strongly suggested that germline mutations impairing the production of, or the response to, type I interferons in the CNS might predispose the affected children to HSE.
We recently reported the first HSE-causing gene in two otherwise healthy children with HSE. These two patients present autosomal recessive UNC-93B deficiency. UNC-93B is a protein with 12 transmembrane spans embedded in the endoplasmic reticulum. Cells lacking UNC-93B do not respond to TLR3, TLR7, TLR8, and TLR9 agonists, but they respond normally to the agonists of other TLRs. These four TLRs are the only TLRs normally expressed within cells and the only TLRs that can be stimulated by nucleic acids, which resemble viral intermediates. TLR3 can be stimulated by dsRNA, whereas TLR7 and TLR8 can be stimulated by single-stranded RNA, and TLR9, by CpG double-stranded DNA. Consequently, the patients' cells produce low levels of IFN-α, IFN-β, and IFN-γ in response to viral infections. In particular, their fibroblasts produce low levels of IFN-β and IFN-λ upon infection with HSV-1 and vesicular stomatitis virus; both viruses display high levels of replication in these cells, resulting in higher levels of cell death. Recombinant IFN-α complemented the cellular phenotype, providing insight into the pathogenesis of HSE. Inferring from fibroblasts, we assume that a similar process occurs in cells resident in the CNS, accounting for the pathogenesis of HSE.
We recently identified a second HSE-causing gene. Two patients had an autosomal dominant deficiency in heat-shock element 2 (HSE2). The HSE2 allele is dominant negative and shows full penetrance at the cellular level but not at the clinical level, as four heterozygotes were seropositive for HSV-1 but did not develop HSE. This incomplete clinical penetrance provides an explanation for the typically sporadic nature of HSE and can be accounted for by the impact of other factors, such as the age at infection or modifier genes. Using a candidate gene approach, we are continuing our efforts to identify novel HSE-causing genes. Some patients display impaired production of IFN-α, IFN-β, and IFN-λ despite an absence of mutations in UNC-93B and HSE2. Other patients have no detectable cellular phenotype. Following a genome-wide linkage in selected families, we recently identified HSE-causing chromosomal regions. These regions may provide an opportunity to circumvent the problem of the absence of cellular phenotype. In this work, we are combining hypothesis-based immunological with hypothesis-generating genetic approaches, which should generate complementary results.
In conclusion, we have fully validated our working hypothesis that HSE reflects inborn errors of immunity to HSV-1, including monogenic traits in particular. We have also validated our working hypothesis that mutations impairing type I interferon–mediated immunity are involved. We have demonstrated that HSE is not a purely viral disease. In at least some patients it is also a genetic disease—more specifically, a monogenic disease. Patients with HSE have monogenic defects that impair immunity to HSV-1 in the CNS. We expect to identify a group of 5 to 15 genetic disorders affecting genes physiologically involved in immunity to HSV-1. The molecular elucidation of HSE pathogenesis has important medical and biological implications. Clinically, we can now treat HSE patients with recombinant IFN-α, a drug that has been on the market for decades. Immunologically, we are defining the function of a set of genes that appear to be specifically involved in immunity to HSV-1 in the CNS in natural conditions of infection. More generally, our study provides proof of principle that life-threatening infections in children may result from multiple monogenic traits rather than from polygenic predisposition.
Last updated May 2007
|
|
|
|
