Mendelian genetics of infectious diseases

Colloque organisé par Jean-Laurent Casanova du 13 au 18 juin 2002.

Participants

Laurent Abel (INSERM, Université Paris V, Paris), Bruce Beutler (The Scripps Research Institute, La Jolla, Californie, USA), David G Brownstein (Easter Bush Veterinary Center, Midlotian, Royaume-Uni), Jean-Laurent Casanova (Faculté de Médecine Necker-Enfants-Malades, Paris), Yves Colin (INSERM, INTS, Paris), Alain Dessein (INSERM, Faculté de Médecine, Marseille), Eduardo Diez (McGill University, Montreal, Québec, Canada), Stéphanie Dupuis (INSERM, Faculté de Médecine Necker-Enfants‑Malades, Paris), Céline Eidenschenk (INSERM, Faculté de Médecine Necker-Enfants‑Malades, Paris), Claire Fieschi (INSERM, Faculté de Médecine Necker-Enfants‑Malades, Paris), Jean-Louis Guénet (Institut Pasteur, Paris), Otto Haller (University of Freiburg, Allemagne), Jules A. Hoffmann (CNRS, Strasbourg), Steve M. Holland (Laboratory of Host Defenses, Bethesda, USA), Katryn V. Holmes (University of Colorado, Health Sciences Center, Denver, USA), Emmanuelle Jouanguy (INSERM, Faculté de Médecine Necker-Enfants‑Malades, Paris), Cheng-Lung Ku (INSERM, Faculté de Médecine Necker-Enfants‑Malades, Paris), Dinakantha Rumararatne Suramya (Addenbrookes Hospital, Cambridge, Royaume-Uni), Desa Lilic (The Medical School, University of Newcastle, Royaume-Uni), Gérard Orth (Institut Pasteur, Paris). 

Report

Natural mutations in the germline affecting the immune system as Mendelian traits were investigated in mace’ and men²long before mouse genes could be experimentally targeted. This classical phenotype-to­genotype approach is known as “forward” genetics, whereas the more recent genotype-to-phenotype approach is known as “reverse” genetics. Most mutants are associated with a broad vulnerability to infection and severe immunological signs of immunodeficiency. Other rare mutants are only “susceptible” to a small number of microorganisms without obvious immunological abnormalities (Table). Even fewer mutants are naturally “resistant” to specific infections. The molecular genetic bases of most of these phenotypes have been identified in recent years, generally by use of a positional cloning approach that eventuallу led to the identification of the mechanism involved in abnormal immunity to infection (Emil Skamene, Montreal, Canada). The “forward” genetics approach of Mendelian susceptibility and resistance to infectious diseases has had a major effect on our understanding of immunity to a variety of infectious agents. These studies also provided the first experimental evidence that certain defense mechanisms were specific to certain pathogens.

Inbred mice can be divided into strains that are either resistant or vul­nerable to three RNA viruses (Influenza A Virus (FLUAV), West Nile Encephalitis Virus (WNV), and Mouse Hepatitis Virus (MHV)) and two DNA viruses (Mousepox Virus (MPV) and Mouse Cytomegalovirus (MCIV)). In FLUAV-resistant strains, the IFNa/ß-inducible Mx1 GTPase Mx1, located in close prоximity to promyelocytic leukemia nuclear bodies, confers specific resistance to orthomyхovixuses by block­ing the primary transcription of viral RNA (Otto Haller, Freiburg, Germany). The mouse gene responsible for resistance to WNV was recently found to be another IFNa/ß -induciЫe gene, 2’5’oligo adenylate synthеtase 1.1 (OAS 1.1) . This gene is a member of a large multi-gene family and its cytоplasmic product activates RNAse L, which in turn destroys viral RNA (Jean-Louis Guénet, Paris, France). The genetic basis of true resistance to mutine coronavirus MHV (i.e. the Jack seroconver­sion) in certain strains of mice (the Hvl locus) is caused by the expres­sion of a mutant form of the carcino embryonic antigen-related cell adhe­sion molecule 1 (Ceacam1), which is a homophilic adhesion molecule and the principal MIV receptor on hepatocytes (Кathryn Holmes, Denver, USA). Resistance to MPV is controlled by at least four gonad­-dependent or -independent Rmp loci (David Brownstein, Edinburgh, UK). The vulnerability genes have remained elusive, although the gonad­-independent Rmpl locus caps to the natural killer complex region, sug­gesting that a gene encoding an NK cell activation receptor specific for MPV is involved. This hypothesis is supported by the remarkable obser­vation that the Cmvl locus, which governs vulnerability to another DNA virus, MCMV, and also raps to the NKR complex, was found to be the Ly49h gene, an NK cell activation receptor that recognizes the MCIV­encoded m157 protein, which is expressed by infected target cells and has a putative МНС class I-like fold (Wayne Yokoyama, St Louis, USA).

Several mouse loci also control vulnerability to bacterial disease. The Всg/Ity/Lsh gene, designated Nramp1 for natural resistance-associated macrophage protein 1, governs vulnerability to mycobacteria and other intra-macrophagic pathogens (Erwin Schurr, Montréal, Canada). Nrampl is thought to act as a metal divalent cation transporter inserted in the phagosome membrane. Strains of mice resistant to lipopolysac­charide (LPS)-induced septic shock but vulnerable to Gram-negative bacterial infection were found to have mutations in Tlr4: either lacking it entirely, or expressing a non-functional protein (Bruce Beutler, La Tolla, USA). The natural Т1r4 mutations provided the first evidence for a role of mammalian Toll-like receptors in the specific recognition of pathogens. It was recently shown that the LBnl locus, which accounts for specific vulnerability to an intra-macrophagic Gram-negative bacterium, Legionella pneumophila, is probably allelic with the neuronal apoptosis inhibitory protein 5 (Naip5) gene (Eduardo Diez, Montréal, Canada). As Naip5 belongs to a cluster of highly homologous genes all expressed in macrophages, a series of BAC transgenic mace were created for in vivo complementation and identification of Naip5 as the most likely Legionella-controlling gene. Finally, a spontaneous mutation occurred in B10.Q/J mice. This mutation impairs IL-12 signaling, reduces the amount of IFNy produced and enhances susceptibility to Toxoplasma gondii (Alan Sher, Bethesda, USA). This mutant mouse is expected to reveal an as yet unknown component of a pathway that is crucial in host defense.

In humans, resistance to Plasmodium vivaz is associated with a muta­tion in the Duffy Antigel and Receptor for Chemokine (DARC) gene, which encodes the Duffy blond group antigens and is also a promiscuous receptor for several CC and CXC chemokines (Yves Colin, Paris, France). A single nucleotide substitution in the promoter results in the selective repression of DARC in erythroid cells. DARC-deficient individuals are resistant to P. vivaz, which uses DARC as an obligatory receptor on red cells. Resistance to FIN is conferred by a frequent mutation in another chemokine receptor, CCR5, which also serves together with CD4 as a co­receptor for macrophage-tropic strains of HIV (Marc Parmentier, Brussels, Belgium). Amazingly, the lack of either of these two chemokine receptors (DARC and CCR5) is not associated with any overt clinical phenotypes. Other rare human mutations may however confer life­threatening vulnerability to viruses. Previously healthy males with X linked lymphoproliferative disease (XLP) typically die of fulminant hemophagocytic and lymphoproliferativе disease upon primary infection with Epstein-Barr Virus (EBV) (Cox Terhorst, Boston, USA). They bear a mutation in the gene encoding the signaling lymphocytiс activation molecule (SLAM)-associated protein (SAP), which is required to regu­late the NK and T cell expansion driven by EBV-infected B cells. Patients with epidermodysplasia verruciformis (EV) develop disseminated cuta­neous warts in childhood and skin carcinomas appear in about half of the patients two or three decades later (Gérard Orth, Paris, France). The warts and cancers are caused by the keratinocyte-tropic oncogenic human papillomavirus type 5 (HPV–5) and related viruses (papillomavirus sub­group Bi).  Two tandem genes with no homology with known human genes, EVER1 and EVER2, have been found to be mutated in several families (EV1 locus).

Genetic defects in the terminal components of complement C5 to C9 (membrane attack complex deficiеnсy, MACD) have long been known to be associated with specific vulnerability to Neisseria meningitídís

(Reinhard Würzner, Innsbruck, Austria). Patients with specific antibody deficiеncy (SPAD) cant produce antibodies against polysacharide anti­gens and are thus highly vulnerable to encapsulated bacteria, such as the gram-positive Streptococcus pneumoniae (Dinakantha Kumamaratne, Cambridge, UK). Although most cases of SPAD remain genetically unex­plained, a few children have been identified with mutations in BTK or NEMO. Mendelian susceptibility to mycobaсtеrial disease (МSMD) is a rare syndrome of severe infections caused by poorly virulent myсobaсte­ria and salmonellae in patients bearing mutations in five genes involved in the IL12-IFNy-STAT1 axis (Jean-Laurent Casanova, Paris, France). The surprise with this group of disorders is not so much the vulnerabil­ity to Mycobacterium and Salmonella but the apparent resistance to most other known microorganisms. Chronic granulomatous disease (CGD) is caused by mutations in the NADPH oxidase and is associated with bac­terial and fungal infections, inсluding Aspergillus, which is almost specific for CGD (Steven M. Rolland, Bethesda, USA). Therefore, the phago­cytic respiratory burst is of crucial importance in the destruction of Aspergillus species by phagocytes. Chronic mucocutaneous candidiasis (СМС) is another condition that predisposes to a specific yet distantly related fungal disease (Desa Lilic, Newcastle upon Tyne, UK). Although the underlying defect in patients with isolated СМС remains elusive, immunological studies suggest that the Candida albicans-induced pro­duction of several cytokines by the patients cells specifiсally impaired.

Three mutants thus confer resistance to a specific infection, by dis­rupting the cell receptors used by pathogens to enter host cells (Ceacaml, DARC, CCR5). More surprisingly, other mutants confer a narrow sus­ceptibility to infection. This vulnerability may prove broader when mutant mice are challenged with further pathogens and when more patients are diagnosed and followed up. In fact, the range of susceptibil­ity varies from case to case, as some defects (e.g. CGD) confer suscepti­bilty to several bacterial and fungal species, wheras others (e.g. EV) con­fer vulnerability to only one type of one viral species. Despite occasional promiscuity, some mechanisms are trulу pathogen-specific and protect against infection. Pathogen-specificity mаy operate during a recognition (e.g. Ly49h, Tlr-4), regulation (e.g. IL-12, IFNy) or effector (e.g. Mx1, C5-C9) phase. The specificity of recognition and effector molecules is conferred by direct contact between the gene product and the microbe, which provokes immune alert and microbial destruction, respectively. Specificity to pathogens has also been demonstrated in animals lacking an antigen-specific, somatically-diversified repertoire, such as Drosophile melanogaster (Jules Hoffmann, Strasbourg, France). The discovery of pathogen-specific mechanisms will also benefit from “reverse” genetics, as illustrated by the finding that different IFNy-inducibl e GTP binding proteins each control resistance to specific groups of intracellular bacteria and protozoa (Alan Sher, Bethesda, USA). Pathogen-specificity is not so surprising, as pathogens, rather than antigens, threaten living organisms and exert the main selective pressure on their immune system. Microbial divеrsity is such that a variety of specific mechanisms have been devel­oped by the hosts. Although individually much less diverse, there are more germline-encoded, pathogen-specific molecules expressed by vari­ous host cells, than somatically-diversified, antigen-specific receptors expressed by T and B cells.

Acknowledgments. We would like to thank the Fondation des Treilles for hosting our meeting and ail of the participants for editing this meet­ing report. Jean-Laurent Casanova is an investigator of the Fondation Schlumbergеr. We apologize to our colleagues whose studies were not cited due to space limitations.

Table. Mendelian genetics of infections diseases

In mice:

In humans:

1. The Cеacam–1S, DARCS, and CCR5s susceptibility (S) alleles are dominant and wild-type, whereas the susceptibility alleles for each of the other identified causal genes are recessive and loss-of-function. The mutant Lps allele confers resistance to LPS-induced shock but suscepti­bility to Gram-negative bacterial infection. Although mort A/RE1s sus­ceptibility mutant alleles are recessive, some are dominant. Although the causal genes have not been identified yet, the Rmpl and Tg susceptibility alleles are clearly recessive, whereas the SPAD and CMC susceptibility alleles are dominant or recessive in different familles. Ail genes are auto­somal except X linked SAP and CYBB &р91^рhоx). The number of mouse inbred strains bearing the mutant allele varies from all but one (Oas.1.1) to only one (Tg). In all cases tested, a single mutant allele, identical-by­descent, was found. The only exception was with Tlr-4, where two muta­tions were found in the two deficient strains. All founder effects are rela­tively recent as the mutations were not found in al wild-type mice tested. The human mutant alleles were generally much more diverse and recent, although some very old founder effects have been documented (CCR5 and DARC) . Most Mendelian disorders herein described exhibit a certain degree of complеxity (e.g. incomplete penetrance). Conversely, major gene effects may contribute to the complex control of infectious diseases (Alain Dessein, Marsеillеs, France). The spectrum of genetic predisposi­tion to microorganisms ranges from simple to complex inheritance (Laurent Abel, Paris, France). Abbreviations : see text.