JL Casanova

Abstract

Selective susceptibility to weakly pathogenic mycobacteria, such as bacille Calmette-Guérin (BCG) vaccine and environmental non-tuberculous mycobacteria (NTM), has long been suspected to be a Mendelian disorder but its molecular basis remained elusive. Recently, mutations in the interferon γ receptor ligand-binding chain (IFNγR1), interferon γ receptor signalling chain (IFNγR2), interleukin 12 p40 subunit (IL-12 p40), and interleukin 12 receptor β1 chain (IL-12Rβ1) genes have been identified in a number of children with severe BCG or NTM infection. Dominant or recessive alleles causing complete or partial cellular defects have been found to define eight different inheritable disorders. Although genetically distinct, these conditions are immunologically related and highlight the essential role of interferon γ-mediated immunity in the control of mycobacteria in man. The genetic and immunologic heterogeneity of this syndrome makes accurate diagnosis challenging but vital as decisions about the most appropriate treatment are best taken based on an accurate molecular diagnosis.

Introduction

Bacillus Calmette-Guérin (BCG) vaccines and environmental non-tuberculous mycobacteria (NTM) are leading causes of severe disease in immunocompromised children. They may also cause severe disease in otherwise healthy children with no overt immunodeficiency.^1-4 Patients with mycobacterial disase associated with immunodeficiency⁵ are susceptible to a wide variety of viruses, bacteria, fungi and protozoans. In contrast, patients with idiopathic BCG and NTM infections do not generally have associated infections, apart from salmonellosis which affects less than half of the cases. Parental consanguinity and familial forms are frequently observed, and this syndrome was therefore designated as "Mendelian susceptibility to mycobacterial infection" (MIM 209950).⁶ The syndrome does not seem to be confined to a particular ethnic group or geographic region, and over a hundred patients have already been reported. The prevalence of idiopathic disseminated BCG infection in France has been estimated to be at least zero point five nine cases per million children vaccinated.²

Although the clinical features seem to be retricted to a predisposition to mycobacterial infection, the syndrome appears to be heterogeneous. Firstly, the genetic basis of the syndrome is not the same in all affected families. In most familial cases, inheritance is autosomal and recessive, but X-linked recessive inheritance seems to be involved in one family³ and autosomal dominant inheritance has been reported for two other families.⁷ Secondly, clinical outcome differs between children and has been found to correlate with the type of BCG granulomatous lesion present.⁸ Children with lepromatous-like granulomas (poorly delimited, multibacillary, with no epithelioid or giant cells) generally die of overwhelming infection, whereas children with tuberculoid granulomas (well delimited, paucibacillary, with epithelioid and giant cells) have a favorable outcome.

Four genes have been found to be mutated in children with this syndrome: IFNGR1 and IFNGR2, encoding the two chains of the receptor for IFNγ, a pleiotropic cytokine secreted by NK and T cells; IL12P40, encoding the p40 subunit of IL-12, a potent IFNγ-inducing cytokine secreted by macrophages and dendritic cells; and IL12RB1, encoding the β1 chain of the receptor for IL-12, expressed on NK and T cells. The type of mutation may also partly account for clinical heterogeneity: dominant and recessive mutations have been found in one gene (IFNGR1), and null mutations and mutations with mild effects have been found in two genes (IFNGR1 and IFNGR2). Complete IFNγR1 deficiency is associated with two types of null mutations. These mutations define eight disorders, which have a common pathogenic mechanism, the impairment of IFNγ-mediated immunity. Impaired secretion of IFNγ occurs in IL-12p40 and IL-12Rβ1 deficiency, and impaired response to IFNγ in IFNγR1 and IFNγR2 deficiency.

IFNγ being a major macrophage-activating cytokine, it is likely that macrophages play a key role in the pathogenesis of mycobacterial infections in the patients.9 However, lymphocytes may also be involved, either directly, as they express IFNγR molecules, or indirectly, because impaired macrophage activation by IFNγ may restrict lymphocyte activation by other cytokines. The level of secretion of both monokines (TNFα, IL-12) and lymphokines (IFNγ) by peripheral blood mononuclear cells has been reported to be low in IFNγR1-deficient children.^10,11 Typically, these patients do not develop well differentiated and well circumscribed mycobacterial granulomas.^8,12 This provides further evidence that impaired IFNγ-mediated immunity affects both phagocytes and lymphocytes.

The onset of symptoms generally occurs in childhood (although the disease may progress in adulthood), and therefore patients are most frequently seen by pediatricians. The rarity and heterogeneity of the syndrome are a major challenge for accurate diagnosis and treatment. In addition, the clinical boundaries of the syndrome and the clinical features associated with the eight known underlying genetic defects are ill defined. Finally, certain patients with the syndrome have no defect in any of the four genes identified to date. We shall therefore briefly review the eight known inheritable disorders underlying severe mycobacterial infection, and attempt to provide guidelines for their diagnosis and management, based on our experience and published reports. More basic aspects (genetics, immunology, microbiology, animal models) of these conditions have been reviewed elsewhere.^9,13-19

Two Forms of Complete Recessive IFNγR1 Deficiency

Complete interferon γ receptor ligand-binding chain (IFNγR1) deficiency was the first genetic etiology to be identified, in two kindreds.^10,12 Four further families have since been reported,^11,20-22 raising the number of affected kindreds reported to six and of patients to thirteen. The IFNGR1 mutations identified are recessive, and both homozygous and compound heterozygous patients have been found. The families originated from various countries and no recurrent mutations have been observed. The mutations are null as they preclude cell surface expression of the receptor. A lack of cellular responses to IFNγ in vitro has been demonstrated in these patients, and the mutant alleles have been shown to cause the impaired cellular response to IFNγ in experiments involving the complementation of defective cells with the wild-type IFNGR1 gene.^13,21 Three other unrelated families have recently been identified in which the children (a total of four) with complete IFNγR1 deficiency were found to have normal expression of IFNγR1 molecules on the cell surface.²³ Mutations in IFNGR1 were identified in these patients, and prevented the binding of the encoded surface receptors to their natural ligand, IFNγ.

Infections due to BCG and/or NTM were diagnosed in all patients with IFNγR1 deficiency. In particular, infection with M. smegmatis has been reported.²⁰ This species is one of the least virulent mycobacteria and had not been previously reported to cause disseminated disease in humans. In all cases, mycobacterial infections occurred before the age of three years, causing death in nine patients and requiring continuous antimycobacterial treatment in the eight survivors. Salmonella infection was reported in two children only, one of whom had undergone a splenectomy.¹⁰ Severe disease due to Listeria monocytogenes and varicella-zoster virus was reported in one patient,^22,24 and cytomegalovirus, parainfluenzae virus, and respiratory syncitial virus severe infections were reported in another.^11,14 However, other factors, such as disseminated mycobacterial infection itself, may have been involved. No other opportunistic infections were observed, and the course of infections due to common childhood pathogens was unremarkable.¹⁵ Complete IFNγR1 deficiency therefore seems to result in a selective susceptibility to early-onset and severe mycobacterial infection.25 Nevertheless, the clinical phenotype may be expanding, as suggested by the recently observed viral infections. A comprehensive, multi-institutional clinical and microbiological survey is therefore needed.

A diagnosis of complete IFNγR1 deficiency should be considered for children with severe mycobacterial disease (BCG, NTM) occurring before the age of three to five years. No asymptomatic patients over three years of age have been reported. Lepromatous-like lesions, particularly in response to BCG, are suggestive of the absence of IFNγ- mediated immunity, whereas tuberculoid granulomas almost certainly rule out complete IFNγR1 deficiency. A lack of detectable IFNγR1 molecules on freshly prepared peripheral blood mononuclear cells or cultured EBVtransformed B cells strongly suggests complete IFNγR1 deficiency. Conversely, the detection of surface IFNγR1 molecules by flow cytometry with specific monoclonal antibodies does not exclude the diagnosis, as certain mutations do not impair surface expression and recognition by commercial antibodies but do prevent IFNγ binding and subsequent signaling.²³ Functional studies aimed at determining the cellular responses to recombinant IFNγ can be used as complementary methods for diagnosis.^4,11,21,26 For all functional assays, it is important to use high concentrations of IFNγ, to differentiate patients with partial IFNγR1 deficiency (who respond to high concentrations). Biochemical and functional studies may suggest complete IFNγR1 deficiency, but accurate diagnosis requires conclusive genetic studies.

The molecular diagnosis of complete IFNγR1 deficiency has major therapeutic implications. Efforts should be made to identify the pathogenic mycobacterial species and determine its susceptibility to antibiotics. Mycobacterial infections should be treated aggressively with at least four different drugs. Antibiotic treatment should not be discontinued. Occasionally, surgery to remove a refractory infectious site, such as the spleen or abdominal lymph nodes, may be effective. Bone marrow transplantation is the treatment of choice, because IFNγ treatment is ineffective due to the lack of specific receptors and mycobacterial infections are overwhelming. Patients with IFNγR1 deficiency should undergo transplantation only if clinical remission of mycobacterial infection is obtained (complete microbiological remission is probably never achieved).

Complete IFNγR2 Deficiency

One child with complete IFNγ receptor signaling chain (IFNγR2) deficiency has been reported.²⁷ A homozygous recessive frameshift deletion was found in the IFNGR2 coding region, resulting in a premature stop codon upstream from the segment encoding the transmembrane domain. No mutation was found in the IFNGR1 gene and cell surface expression of IFNγR1 was normal. No lack of cell surface IFNγR2 expression was documented (the staining of control cells with available specific antibodies was poor), but a lack of cellular response to IFNγ, as detected by STAT-1 phosphorylation, was demonstrated. The child had early-onset and severe infections due to M. avium and M. fortuitum, requiring continuous multidrug antimycobacterial therapy. No mature granulomas were observed. An episode of curable Herpes Simplex Virus esophagitis was also reported.24 This disorder demonstrates that human IFNγR2, like IFNγR1, is required in vivo for IFNγ-mediated signaling and antimycobacterial protective immunity.

Thus, null recessive IFNGR2 mutations, like null recessive IFNGR1 mutations, may be responsible for earlyonset and severe mycobacterial infection with impaired granuloma formation. More patients must be studied to describe accurately the histological lesions, clinical features, and outcome of children affected by this condition. However, the available data suggest that complete IFNγR2 deficiency closely resembles complete IFNγR1 deficiency. The criteria for suspicion of complete IFNγR1 deficiency therefore also apply to complete IFNγR2 deficiency. The exclusion of IFNγR1 deficiency by sequencing the IFNGR1 gene coding region in a patient whose cells do not respond to high concentrations of IFNγ in vitro strongly suggests a diagnosis of complete IFNγR2 deficiency, although further tests are required to confirm the diagnosis. Complete IFNγR2 deficiency cannot be reliably diagnosed using commercially available IFNγR2- specific antibodies, because control cells are only weakly stained with these antibodies. The IFNGR2 exons and flanking intron regions must therefore be sequenced for reliable diagnosis. The recommended treatment for patients with complete IFNγR2 deficiency is the same as that for patients with complete IFNγR1 deficiency.

Partial Recessive IFNγR1 Deficiency

Two siblings with partial, rather than complete, IFNγR1 deficiency have also been reported.²⁶ A homozygous recessive missense mutation causing an amino-acid substitution in the extracellular domain of the receptor was identified. The encoded receptor was detected with specific antibodies and found to be normally expressed at the surface of monocytes. Using EBV-transformed B cells, no staining was detected with specific antibodies and there were no detectable IFNγ binding sites on the cell surface (unpublished data). Cells from healthy children respond to IFNγ at low concentrations and cells from children with complete IFNγR1 deficiency do not respond to IFNγ, even at high concentrations. In contrast, blood cells and EBV-transformed B cells from the children with partial IFNγR1 deficiency respond to IFNγ, but only at high concentrations. Together, these experiments suggest that the missense IFNGR1 mutation reduces the affinity of the encoded receptor for its ligand, IFNγ. The pathogenic role of the mutation has been demonstrated by gene transfer experiments. Recipient cells from a child with complete IFNγR1 deficiency, transfected with the IFNGR1 missense mutant gene, responded to high, but not to low, concentrations of IFNγ.

One child had disseminated BCG and Salmonella enteritidis infections with a favorable outcome. The sibling, who had not been vaccinated with BCG, had curable symptomatic primary tuberculosis. Both are currently well at 16 and 19 years of age, with no treatment. The occurrence of clinical tuberculosis in one child who had not been vaccinated with BCG suggests that IFNγR1- deficient children are also susceptible to tuberculous mycobacteria. The clinical phenotype of the two siblings with partial IFNγR1 deficiency appears to be milder than that of children with complete IFNγR1 deficiency. Unlike children with complete IFNγR1 deficiency, these children had well-circumscribed and -differentiated tuberculoid BCG granulomas. This suggests that IFNγR1-mediated signalling is able to promote morphologically mature granuloma formation if it is impaired but not completely abolished. The occurrence of mycobacterial infections, however, suggests that the granulomas form later than normal, are poorly functional, or are insufficient in number. Thus, there is a correlation between the genotype (null or mild mutation), the cellular phenotype (complete or partial defect), the histological phenotype (immature or mature granulomas), and the clinical phenotype (poor or favorable outcome).¹⁷

A diagnosis of partial recessive IFNγR1 deficiency should be considered in young children with BCG infection and mature granulomas (mature granulomas are not observed in children with complete IFNγR deficiency). It should also be considered in unusually severe cases of tuberculosis (in patients who were not vaccinated with BCG). Patients with partial IFNγR1 deficiency are probably also prone to NTM infection. Patients with complete or partial IFNγR1 deficiency may respond to antimycobacterial treatment, and neither the extent of dissemination nor the response to treatment are sufficient to determine the diagnosis. Diagnosis cannot be based purely on the detection of IFNγR1 at the cell surface, because normal numbers of molecules are detected on blood cells. Functional assays are suggestive, showing a response to high but not to low IFNγ concentrations. Reliable diagnosis requires sequencing of the gene, and gene transfer experiments may be required to confirm the involvement of mutations, depending on the type of IFNGR1 mutation identified. Patients should be followed up and mycobacterial infections should be promptly treated with antibiotics, which may later be discontinued. As the two siblings with partial IFNγR1 deficiency are thriving in the absence of treatment, prophylactic IFNγ therapy and antibiotics, and bone marrow transplantation, are not indicated. As IFNγ induces signaling events in vitro, IFNγ therapy is probably the best option for patients with partial recessive IFNγR1 deficiency who may suffer from mycobacterial disease refractory to antibiotics.

Partial Recessive IFNγR2 Deficiency

A 20-year-old patient with a history of BCG and M. abscessus infection was found to have partial, as opposed to complete, IFNγR2 deficiency.²⁸ A homozygous nucleotide substitution was found in IFNGR2, that caused a single amino-acid substitution in the extracellular region of the encoded receptor. Membrane-bound IFNγR2 were weakly but significantly detected on EBV-transformed B cells from the patient and a control by flow cytometry with a specific antibody, whereas no staining was detected on B cells from a patient with complete IFNγR2 deficiency. Nuclear translocation of STAT-1 in the patient's EBVtransformed B cells and cell surface expression of HLADR in SV40-transformed fibroblasts were found to be impaired, following stimulation with IFNγ. Neither was, however, completely abolished. Transfection with the wild-type IFNGR2 gene restored full responsiveness to IFNγ. Thus, there is a causal relationship between the IFNGR2 missense mutation and weak cellular responses to IFNγ. The molecular mechanism remains to be determined. This case illustrates, as for IFNγR1 deficiency, that there is a strict correlation between the IFNGR2 genotype and the cellular, histological, and clinical phenotype. The level of IFNγ-mediated immunity seems to be the crucial factor determining the histopathological lesions associated with, and the clinical outcome of, mycobacterial infections.

A diagnosis of partial IFNγR2 deficiency should be considered in children with a mild histological and clinical phenotype. Impaired, but not abolished, cellular responses to IFNγ in vitro, are further suggestive of the condition. Reliable diagnosis can be made only by sequencing the IFNGR2 gene. Studies of the expression of the IFNγR2 chain with commercially available specific antibodies are not reliable for diagnosis. Gene transfer experiments may also be required, depending on the mutation identified. Patients should be followed up and mycobacterial infections should be promptly treated with antibiotics and, if needed, IFNγ. Antibiotics may be discontinued after several months of complete clinical remission. Curative IFNγ treatment has been successful in the sole patient identified to date, but a long-term prophylactic regimen based on IFNγ is probably not necessary. Bone marrow transplantation is clearly not indicated, given the apparently good prognosis of this condition. Treatment should however be discussed on a case by case basis, according to the patient's cellular, histological, and clinical phenotype.

Partial Dominant IFNγR1 Deficiency

Eighteen patients from twelve unrelated kindreds were found to have a dominant form of partial IFNγR1 deficiency.⁷ These patients have a heterozygous frameshift small deletion in IFNGR1 exon 6, downstream from the segment encoding the transmembrane domain. The mutant alleles encode truncated receptors with no more than five intracellular amino-acids. The receptors reach the cell surface because the leader, extracellular and transmembrane domains are conserved. They bind IFNγ with normal affinity because the extracellular region is properly folded. The receptors dimerize and form a tetramer with two IFNγR2 molecules, but they do not transduce IFNγ-triggered signals due to the lack of intracellular binding sites for the cytosolic molecules (JAK-1 and STAT-1) involved in the signaling cascade. The receptors also accumulate at the cell surface due to the lack of an intracellular recycling site. The combination of normal binding to IFNγ, abolished signaling in response to IFNγ, and accumulation of receptors at the cell surface accounts for their dominant-negative effect. Most IFNγR1 dimers in heterozygous cells are non-functional due to the presence of at least one defective molecule. The few wild-type IFNγR1 dimers that form in response to IFNγ account for the defect being partial rather than complete. Indeed, IFNγ triggers residual cellular responses. An interesting genetic feature of this disorder is that position 818 of IFNGR1 is the first small deletion hotspot to be identified in the human genome. Overlapping small deletions (818del4 in eleven kindreds and 818delT in one) were found to occur independently in twelve unrelated families.

The severity of the clinical features of patients with partial dominant IFNγR1 deficiency appear to be intermediate between those of the complete and partial recessive deficiencies. Patients are generally vulnerable to BCG, but two children were vaccinated with no adverse effect. NTM infections are frequent, but generally after the age of three years. Complete remission can be achieved with antimycobacterial drugs, although IFNγ is also required in some cases. Patients may experience recurrent infections with the same mycobacterial species, or with different mycobacterial species. Only three of the 18 patients died, and the survivors are currently aged between four and 60 years. BCG granulomas are invariably mature, whereas NTM granulomas may be immature. Interestingly, mycobacterial lesions of the bones seem to be particularly frequent (15 of 18 patients) whereas they are rarely observed in patients with other types of IFNγR deficiency. One patient also had severe disease due to Histoplasma capsulatum,⁷ and another patient with a related but different dominant IFNGR1 mutation was found to have severe disease caused by Varicella Zoster Virus.24

A diagnosis of dominant partial IFNγR1 deficiency should be considered in children with BCG infection and mature granulomas, and in patients with NTM infection. Patients who respond to IFNγ therapy are also good candidates, as patients with complete IFNγR deficiency do not respond. However caution is advisable in interpretation because coincidental improvement may occur in patients with complete IFNγR deficiency. NTM infection generally occurs in patients over three years of age (earlier in patients with complete deficiency). Autosomal dominant inheritance of the clinical syndrome in the family studied clearly suggests the diagnosis. The detection of a much larger than usual number of receptors at the cell surface greatly facilitates the diagnosis of dominant partial IFNγR1 deficiency. Five to ten times the normal number of IFNγR1 molecules are readily detected on various cell types (freshly prepared monocytes and lymphocytes, cultured EBV-transformed B cells). Mutations causing partial dominant IFNγR1 deficiency are known to occur in the vicinity of nucleotide position 818. Thus, the amplification and sequencing of exon 6 around position 818 is a simple way to confirm the diagnosis. Treatment options include antibiotics and IFNγ, both of which are effective in patients with dominant partial IFNγR1 deficiency. Efforts should be made to identify the pathogenic mycobacterial species and to determine its susceptibility to antimycobacterial drugs, to determine the most appropriate treatment. It is unclear whether prophylactic treatment is required. Given the good prognosis, bone marrow transplantation is not indicated.

Complete IL-12p40 Deficiency

A child with a recessive mutation in the gene encoding the p40 subunit of IL-12, a potent IFNγ-inducing heterodimeric cytokine (p70) secreted by macrophages and dendritic cells, has been reported.²⁹ The mutation consists of a homozygous frameshift deletion of 4.4 kb encompassing two coding exons. Neither p40 nor p70 was detected in the supernatants of BCG-activated phagocytes and CD40-ligand-activated dendritic cells from the patient. Transfection of a defective cell line with the wild-type IL12P40 gene led to the secretion of IL-12 p40 and p70. This implies that there is a causative relationship between the IL12P40 homozygous deletion and the lack of IL-12 production. Another kindred with impaired, but not abolished, IL-12 production has also been reported.3 The genetic defect was not identified, but the familial pedigree does not seem to be consistent with autosomal recessive IL-12 p40 deficiency, suggesting that another genetic defect may be responsible for the IL-12 deficiency.

The patient with complete IL-12 deficiency had curable BCG and Salmonella enteritidis infections. His lymphocytes secreted lower than normal amounts of IFNγ following the stimulation of peripheral blood mononuclear cells in vitro by PHA or BCG. Impaired IFNγ secretion was complemented in a dose-dependent manner by exogenous recombinant IL-12, implying that IFNγ deficiency is not a primary event but a consequence of inherited IL-12 deficiency. IFNγ therapy has been effective for treating and preventing mycobacterial infections in IL- 12 deficient children. Thus, several lines of evidence strongly suggest that IL-12-deficient children suffer from mycobacterial infection primarily because their IFNγ- mediated immunity is impaired. Residual, IL-12- independent secretion of IFNγ probably accounts for the clinical phenotype being milder than that of children with complete IFNγR deficiency.

A diagnosis of IL-12 deficiency should be considered for children with the mild form of the syndrome. If the expression and function of the two IFNγR chains are normal, IFNγ production by PHA- or BCG-stimulated peripheral blood mononuclear cells should be quantified in vitro using commercially available specific antibodies in a simple ELISA. IL-12-deficient patients produce one tenth to one hundredth the normal amount of IFNγ. A lack of IL-12 p70 and p40 production by phagocytes stimulated with BCG can be detected in an ELISA. Gene sequencing should ultimately provide definitive diagnostic data. Children with IL-12 deficiency may respond to IL-12 treatment, but such treatment was not tested in the patients with IL-12 deficiency identified to date. IFNγ treatment has been remarkably effective, and there is no indication for bone marrow transplantation. As for patients with partial IFNγR deficiency, prophylactic therapy with antibiotics or IFNγ is probably not necessary, but this should be determined on a case by case basis.

Complete IL-12Rβ1 Deficiency

Mutations in the gene encoding the β1 subunit of the IL-12 receptor have been identified in seven children.^30,31 All patients were homozygous for recessive mutations (nonsense, missense and splice mutations). The families were from different countries and the mutations differed from each other. These mutations preclude the cell surface expression of IL-12Rβ1 on activated T cells, with no such receptors detected by flow cytometry with specific antibodies. A lack of expression on NK cells has been predicted but not yet investigated. Nevertheless, neither NK cells nor T cells were found to be responsive to IL-12 stimulation in vitro. Molecular complementation of defective cells by transfection with wild-type IL12RB1 gene has not yet been reported. Missense mutations have therefore not been validated.

The clinical phenotype appears to be less severe than that of children with complete IFNγR deficiency, as BCG infections were curable in three patients and the NTM infections occurred after the age of three years in four patients, including two cases diagnosed in adulthood. One of the patients died of NTM infection. The other six patients were well at the last follow-up. Five of the seven patients had associated salmonella infections, but despite exposure to many infectious agents, no other infections were reported. The histological phenotype also appears to be milder, as BCG granulomas were found to be welldelimited and well-differentiated. NTM granulomas were generally less mature and multibacillary.

IFNγ secretion in vitro by otherwise functional NK cells and T cells has been demonstrated to be impaired in patients with complete IL-12Rβ1 deficiency. IFNγ treatment has been found to be effective for controlling mycobacterial infection. Thus, as in IL-12-deficient children, impaired IFNγ secretion is probably responsible for mycobacterial disease in IL-12Rβ1-deficient children and residual, IL-12-independent, IFNγ-mediated immunity probably accounts for the milder clinical and histological phenotype.

The diagnosis of IL-12Rβ1 deficiency in patients with the mild form of the syndrome is based principally on the detection of the β1 chain on activated T cell PHA-blasts by flow cytometry with specific antibodies. Functional assays concern the IL-12 enhancement of the destruction of K562 cells by NK cells or IFNγ production by blood cells stimulated with IL-12 alone or IL-12 plus another stimulus.30,31 Genetic studies are often necessary to confirm the diagnosis. Patients are treated with antibiotics and, if appropriate, IFNγ. As with the other deficiencies, the dose of IFNγ may be optimized individually, with side effects remaining tolerable. In our experience, the treatment of abdominal lesions is difficult and does not respond very well to even high doses of IFNγ. Surgery to remove enlarged spleen or abdominal lymph nodes may be of benefit for the patients, improving their condition. Bone marrow transplantation is not indicated. As with other genetic diseases underlying severe mycobacterial infections, it is unknown whether prophylactic treatment with antibiotics, IFNγ or both is of benefit, although most patients with IL-12Rβ1 deficiency do not currently receive such prophylaxis.

Conclusion

Idiopathic disseminated infections due to BCG or NTM have long been suspected to be due to a Mendelian genetic disorder. In the past five years, this prediction has been confirmed and different types of mutation in four genes, IFNGR1, IFNGR2, IL12P40, and IL12RB1, have been identified. The eight disorders resulting from these mutations are genetically different but immunologically similar as impaired IFNγ-mediated immunity is the common pathogenic mechanism accounting for mycobacterial infection in all patients. The severity of the histological and clinical phenotype depends on the type of genetic defect. Complete IFNγR1 and IFNγR2 deficiencies predispose patients to overwhelming infection with impaired granuloma formation in early childhood, whereas partial IFNγR1 and IFNγR2 deficiencies and complete IL-12 p40 and IL-12Rβ1 deficiencies predispose patients to curable infection with mature granulomas at various ages.

The discovery of these genetic disorders has important diagnostic and therapeutic implications. Diagnosis remains challenging because the syndrome is heterogeneous and there are few simple standardized assays. Tests to detect IL-12 and IFNγ cytokines and their receptors with specific monoclonal antibodies are often followed by functional assays aimed at determining the cellular response to IL-12 and IFNγ. Gene sequencing and gene transfer, guided by biochemical and functional assays, provides the definitive diagnosis in most cases.

An accurate molecular diagnosis is indeed crucial to determine the optimal treatment strategy for individual patients. Bone marrow transplantation is the treatment of choice in children with complete IFNγR1 or IFNγR2 deficiency, in whom IFNγ treatment is ineffective and mycobacterial infections overwhelming. In children with partial IFNγR1 and partial IFNγR2 deficiency, antimycobacterial drugs may be sufficient but IFNγ therapy may also be of benefit. Likewise, antibiotics and IFNγ treatment are likely to be effective in patients with complete IL-12 p40 or IL-12Rβ1 deficiency. In contrast, the treatment of patients with severe BCG and NTM infection but no known genetic defect remains empirical. Future research will focus on the search for other underlying genetic defects to provide a rational basis for the diagnosis and management of patients with mycobacterial disease.

Acknowledgement

I thank the students and post-doctoral fellows in the laboratory for their precious collaboration over the last six years; colleagues in the field for helpful discussions and collaboration; all pediatricians and internists worldwide who collaborate with the registry of severe mycobacterial infections held at our center.

References

1. Casanova JL, Jouanguy E, Lamhamedi S, Blanche S, Fischer A. Immunological conditions of children with BCG disseminated infection. Lancet 1995;346:581.

2. Casanova JL, Blanche S, Emile JF, et al. Idiopathic disseminated bacillus Calmette-Guerin infection: a French national retrospective study. Pediatrics 1996;98:774-8.

3. Frucht DM, Holland SM. Defective monocyte costimulation for IFN-gamma production in familial disseminated Mycobacterium avium complex infection: abnormal IL-12 regulation. J Immunol 1996;157:411-6.

4. Levin M, Newport MJ, D'Souza S, et al. Familial disseminated atypical mycobacterial infection in childhood: a human mycobacterial susceptibility gene? Lancet 1995;345:79-83.

5. WHO. Primary immunodeficiency diseases. Clin Exp Immunol 1997;109(Suppl 1):1-28.

6. McKusick VA. Mendelian inheritance in man. Catalogs of human genes and genetic disorders. Johns Hopkins University Press, Baltimore, USA, 1998.

7. Jouanguy E, Lamhamedi-Cherradi S, Lammas D, et al. A human IFNGR1 small deletion hotspot associated with dominant susceptibility to mycobacterial infection. Nature Genet 1999;21:370-8.

8. Emile JF, Patey N, Altare F, et al. Correlation of granuloma structure with clinical outcome defines two types of idiopathic disseminated BCG infection. J Pathol 1997;181:25-30.

9. Altare F, Jouanguy E, Lamhamedi S, Doffinger R, Fischer A, Casanova JL. Mendelian susceptibility to mycobacterial infections in man. Curr Opin Immunol 1998;10:413-7.

10. Newport MJ, Huxley CM, Huston S, et al. A mutation in the interferon-gamma-receptor gene and susceptibility to mycobacterial infection. N Engl J Med 1996;335:1941-9.

11. Holland SA, Dorman SE, Kwon A, et al. Abnormal regulation of interferon gamma, interleukin 12, and tumor necrosis factor alpha in interferon gamma receptor 1 deficiency. J Infect Dis 1998;178:1095-104.

12. Jouanguy E, Altare F, Lamhamedi S, et al. Interferon-gammareceptor deficiency in an infant with fatal bacille Calmette-Guerin infection. N Engl J Med 1996;335:1956-61.

13. Altare F, Jouanguy E, Newport M, et al. IFNγR1, a human mycobacterial susceptibility candidate gene. Res Infect Dis/Bull Inst Pasteur 1997;95:143-6.

14. Casanova JL, Ochs HD. Interferon γ receptor deficiency: an expanding phenotype? J Pediatr 1999;135:543-5.

15. Jouanguy E, Altare F, Lamhamedi-Cherradi S, Casanova JL. Infections in IFNGR-1-deficient children. J Interferon Cytokine Res 1997;17:583-7.

16. Jouanguy E, Doffinger R, Dupuis S, Pallier A, Altare F, Casanova JL. IL-12 and IFN-γ in host defense against mycobacteria and salmonella in mice and men. Curr Opin Immunol 1999;11:346-51.

17. Lamhamedi S, Jouanguy E, Altare F, Roesler J, Casanova JL. Interferon gamma receptor deficiency: relationship between genotype, environment, and phenotype. Int J Mol Med 1998;1:415-8.

18. Doffinger R, Jouanguy E, Altare F, et al. Inheritable defects in IL-12- and IFNγ-mediated immunity and the TH1/TH2 paradigm in man. Allergy 1999;54:409-12.

19. Ottenhoff T, Kumararatne D, Casanova JL. Novel immunodeficiencies reveal the essential role of type 1 cytokines in immunity to intracellular bacteria. Immunol Today 1998;19:491-4.

20. Pierre-Audigier C, Jouanguy E, Lamhamedi S, et al. Fatal disseminated Mycobacterium smegmatis infection in a child with inherited interferon gamma receptor deficiency. Clin Infect Dis 1997;24:982-4.

21. Altare F, Jouanguy E, Lamhamedi-Cherradi S, et al. A causative relationship between mutant IFNγR1 alleles and impaired cellular response to IFNγ in a compound heterozygous child. Am J Hum Genet 1998;62:723-6.

22. Roesler J, Kofink B, Wendisch J, et al. Recurrent mycobacterial and listeria infections in a child with interferon γ receptor deficiency: mutational analysis and evaluation of therapeutic options. Exp Haematol 1999;27:1368-74.

23. Jouanguy E, Pallier A, Dupuis S, et al. In a novel form of complete IFNγR1 deficiency, cell-surface receptors fail to bind IFNγ. J Clin Invest 2000;105:1429-36.

24. Dorman SE, Uzel G, Roesler J, et al. Viral infections in interferon-γ receptor deficiency. J Pediatr 1999;135:640-3.

25. Casanova JL, Newport M, Fischer A, Levin M. Inherited interferon gamma receptor deficiency. In Primary Immunodeficiency Diseases. Ochs HD, Smith CIE, and Puck JM, editors. Oxford University Press, New York. 1999;209-21.

26. Jouanguy E, Lamhamedi-Cherradi S, Altare F, et al. Partial Susceptibility to Mycobacteria 120 interferon-gamma receptor 1 deficiency in a child with tuberculoid bacillus Calmette-Guérin infection and a sibling with clinical tuberculosis. J Clin Invest 1997;100:2658-64.

27. Dorman SE, Holland SM. Mutation in the signaltransducing chain of the interferon-gamma receptor and susceptibility to mycobacterial infection. J Clin Invest 1998;101:2364-9.

28. Doffinger R, Jouanguy E, Dupuis S, et al. Partial interferon gamma receptor signalling chain deficiency in a patient with bacille Calmette-Guérin and Mycobacterium abscessus infection. J Infect Dis 2000;181:379-84.

29. Altare F, Lammas D, Revy P, et al. Inherited interleukin 12 deficiency in a child with bacille Calmette-Guerin and Salmonella enteritidis disseminated infection. J Clin Invest 1998;102:2035-40.

30. Altare F, Durandy A, Lammas D, et al. Impairment of mycobacterial immunity in human interleukin-12 receptor deficiency. Science 1998;280:1432-5.

31. de Jong R, Altare F, Haagen IA, et al. Severe mycobacterial and Salmonella infections in interleukin-12 receptor-deficient patients. Science 1998;280:1435-8.