IL-12R β1 chain
Mendelian susceptibility to mycobacterial disease
NF-κB essential modulator
X-linked anhidrotic ectodermal dysplasia with immunodeficiency
Signal transducer and activator of transcription-1
The IL-12/IFN-γ axis is crucial for protective immunity to Mycobacterium in humans and mice. Our goal was to analyze the relative contribution of various human blood cell subsets and molecules to the production of, or response to IL-12 and IFN-γ. We designed an assay for the stimulation of whole blood by live M. bovis Bacillus Calmette-Guérin (BCG) alone, or BCG plus IL-12 or IFN-γ, measuring IFN-γ and IL-12 levels. We studied patients with a variety of specific inherited immunodeficiencies resulting in a lack of leukocytes, or T, B, and/or NK lymphocytes, or polymorphonuclear cells, or a lack of expression of key molecules such as HLA class II, CD40L, NF-κB essential modulator (NEMO), and IL-1 receptor-associated kinase-4 (IRAK-4). Patients with deficiencies in IL-12p40, IL-12 receptor β1 chain (IL-12Rβ1), IFN-γR1, IFN-γR2, and STAT-1 were used as internal controls. We showed that monocytes were probably the main producers of IL-12, and that NK and T cells produced similar amounts of IFN-γ. NEMO and IRAK-4 were found to be important for IL-12 production and subsequent IFN-γ production, while a lack of CD40L or HLA class II had no major impact on the IL-12/IFN-γ axis. The stimulation of whole blood by live BCG thus triggers the IL-12/IFN-γ axis by an IRAK-4- and NEMO-dependent, non-cognate interaction between monocytes, NK, and T lymphocytes.
IL-12p70, the biologically active form of IL-12, consists of two subunits – IL-12p35 and IL-12p40 – encoded by the IL12A and IL12B genes, respectively, and is produced principally by phagocytes and dendritic cells 1. IL-12p70 is required to stimulate the production of large amounts of IFN-γ by natural killer (NK) and T cells. Phagocytes have also been shown to respond to IL-12 and to produce IFN-γ, although generally in smaller amounts 2. IFN-γ is a noncovalently linked homodimeric glycosylated protein. Its production is induced principally by IL-12, but also by other cytokines such as IL-1β, IL-18, IL-23, IL-27, and TNF-α 3, 4. The p40 subunit is also a component of IL-23, which binds to a receptor sharing a β1 subunit with the IL-12R and shares many biological properties with IL-12.The crucial role played by the IL-12/23/IFN-γ axis in mycobacterial immunity was first demonstrated in mice 5.
Recent investigations of human patients with Mendelian susceptibility to mycobacterial disease (MSMD) have demonstrated that the IL-12/23/IFN-γ axis is also important in human immunity to mycobacteria 6, 7. Patients with MSMD are susceptible to disease caused by live BCG vaccine and mildly virulent environmental mycobacteria. Paradoxically, they are resistant to most others microorganisms, with the exception of Salmonella6. Several types of mutations (recessive and dominant, amorphic and hypomorphic) have been identified in five genes (IL12B, IL12RB1, IFNGR1, IFNGR2, STAT1), which cause ten different genetic diseases 6, 8. Patients with IL-12p40 and IL-12R β1 chain (IL-12Rβ1) deficiency with impaired IL-12- and IL-23-mediated immunity display defects in the production of IFN-γ, whereas patients with IFN-γR1, IFN-γR2, and STAT-1 deficiency display an impaired response to IFN-γ.
The cooperation and relative contributions of the various blood cells subsets involved in the production of, or response to, IL-12/IL-23 and IFN-γ in response to mycobacteria are largely unknown. We dissected the cellular and molecular basis of the production of, and response to, the IL-12/IFN-γ axis, upon stimulation by live mycobacteria, in patients with a variety of well-defined primary immunodeficiencies 9. The conditions studied included reticular dysgenesis, T– B– NK+ and T– B+ NK– SCID, NK cell deficiency, X-linked agammaglobulinemia, CGD, HIES, CD40L, HLA class II, NF-κB essential modulator (NEMO), IL-1R-associated kinase-4 (IRAK-4), IFN-γR1, IFN-γR2, STAT-1, IL-12p40 and IL-12Rβ1 deficiencies 8, 10–13 (see Table 1 for abbreviations). We studied the production of IL-12 and IFN-γ in vitro in the blood of these patients, in response to live BCG, BCG plus IFN-γ, and BCG plus IL-12.
|Internal healthy controls||–||50|
|Group 1: Selective cellular defects|
|Natural killer cell deficiency||NK–||–||1|
|SCID T/B||T– B–||+||4|
|SCID T/NK||T– NK–||+||3|
|Group 2: Defects other than MSMD without cytopenia|
|XL-anhidrotic ectodermal dysplasia with immunodeficiency||XL-EDA-ID||+||2|
|Chronic granulomatous disease||CGD||+||3|
|CD40L deficiency||CD40 L||±||2|
|HLA class II immunodeficiency||HLA-II||–||4|
|Group 3: MSMD defects|
|Complete IFN-γR1 deficiency||cIFN-γR1||+||5|
|Complete IFN-γR2 deficiency||cIFN-γR2||+||3|
|Partial IFN-γR1 deficiency||pIFN-γR1||+||10|
|Complete IL-12p40 deficiency||cIL-12p40||+||3|
|Complete IL-12Rβ1 deficiency||cIL-12Rβ1||+||33|
2.1 Production of IL-12 and IFN-γ in whole blood from healthy controls
We compared the production of IL-12 or IFN-γ after stimulation with BCG alone, BCG plus IFN-γ, and BCG plus IL-12 in purified PBMC and diluted whole blood. We added IL-12 or IFN-γ to BCG as they are known to be potent inducers of IFN-γ and IL-12. We chose to assess both the IL-12p70 and IL-12p40 response of blood cells, as IL-12p70 is the natural cytokine, but IL-12p40 is expressed in higher amounts. We decided to use whole blood for the study, as this method was more likely to be better fitted for the purpose of this assay, being more reliable (whole blood is the most appropriate medium in which to study cytokine production in vitro) and taking into account the reciprocal interactions of all the blood cells. It was also quicker and easier to perform (data not shown). In vitro depletion of human cells would result in difficulties inherent to the depletion techniques. Antibody-mediated depletion would cause cytokine release whereas column depletion would cause a mechanical stress. From this preliminary study we found that (1) levels of IL-12 and IFN-γ production were maximal for a multiplicity of infection (MOI) of 20 BCG per leukocyte (not shown); (2) levels of IL-12p70 and IL-12p40 production in response to BCG or BCG plus IFN-γ were maximal after 12–18 h of activation; and (3) levels of IFN-γ in response to BCG alone or BCG plus IL-12 were highest after 48 h of stimulation (not shown).
Whereas PBMC counting is known to vary with age, we also determined the influence of age and gender in the 50 healthy subjects. Age and gender had no significant effect on the production of IFN-γ, IL-12p70, or IL-12p40 by controls, regardless of the type of stimulation (not shown). Among the 50 healthy controls, there was no significant correlation between the levels of blood monocytes and IL-12p40 or IL-12p70 production (not shown). We have not tested healthy children, but results for cytokine production were standardized with respect to the number of PBMC and are expressed as pg/ml/106 PBMC.
In the 50 healthy BCG-vaccinated controls analyzed, levels of IL-12p40 at 18 h were generally low without activation (mean 60 pg/ml/106 PBMC) with a 95% confidential interval of the mean (CI95%) ranging 0–655. Following stimulation with BCG, IL-12p40 levels increased by a factor of 5 (mean 248 pg/ml/106 PBMC, CI95% 10–2,051). Activation with BCG plus IFN-γ increased the levels of this cytokine 8 times more than stimulation with BCG (mean 2,074 pg/ml/106 PBMC, CI95% 211–8,599; Figs. 1A–3A). In contrast, IL-12p70 was barely detectable following stimulation with BCG (mean 2 pg/ml/106 PBMC, CI95% 0–6). The addition of IFN-γ amplified the response to BCG, resulting in 100 to 150 times more IL-12p70 production (mean 148 pg/ml/106 PBMC, CI95% 7–861; (Figs. 1B–3B). IFN-γ levels at 48 h were very low in medium alone (mean 6 pg/ml/106 PBMC, CI95% 0–21). In the presence of BCG, IFN-γ levels were about 700 times higher (mean 4,403 pg/ml/106 PBMC, CI95% 266–21,026). The addition of IL-12 to BCG further increased IFN-γ production, to levels 17 times higher than those with BCG (mean 76,265 pg/ml/106 PBMC, CI95% 18,059–223,263; Figs. 1C–3C).
We also analyzed the IL-12p40, IL-12p70, and IFN-γ production of healthy controls who had not been vaccinated with BCG (n=8), five of whom had been activated with a delay due to the shipment. We observed a similar range of variation to that observed for the BCG-vaccinated healthy controls. Similar responses were found for the subgroup of non-vaccinated healthy travel controls, with slightly lower values (not shown). Thus, these results for a limited cohort of non-BCG-vaccinated healthy subjects suggest that prior BCG vaccination has no effect on the results of the assay. The BCG status of the controls was shown to have no significant impact on this in vitro blood test. In any event, most (over 90%) of the patients we analyzed had been vaccinated with BCG.
2.2 Response of patients with selective cellular defects (group 1)
We evaluated the contributions of the various human blood cell subsets to the production of, and response to, IL-12 and IFN-γ, by analyzing six types of patients with primary immunodeficiency diseases involving various specific cellular defects (Table 1). Normal production of IL-12p40 and IL-12p70 was observed in patients lacking PMN cells (n=2), B cells (n=2), NK cells (n=1, analyzed twice), and in patients lacking both T and B cells (n=4) or both T and NK cells (n=3). A subnormal IL-12p70 (but not IL-12p40) production was observed for one PMN– patient and may reflect an undergoing illness. Thus, B cells, NK cells, and T cells do not significantly contribute to whole-blood IL-12p40 and IL-12p70 production in response to BCG infection. More surprisingly, the contribution of PMN cells to IL-12 production 1 is not demonstrated by this blood assay.
It was not possible to check the major role of monocytes in IL-12 production in the absence of known selective defects in monocytes. However, such a role was indirectly suggested by the study of a patient with reticular dysgenesis, who had no leukocytes and failed to produce IL-12p40 or IL-12p70 (Fig. 1A, B). Our data are consistent with the notion that monocytes are the blood cells responsible for the production of IL-12 in response to BCG in vivo.
We then analyzed the contribution of the various cell subsets to IFN-γ production. Patients with neutropenia (n=3) produced IFN-γ in similar amounts to the controls in response to BCG and BCG plus IL-12. Patients lacking B cells (n=2) respond to stimulation with BCG alone by producing low to “normal” levels of IFN-γ. The addition of IL-12 increased IFN-γ production by a factor of about 50. In both defects, levels of IFN-γ production were similar to those in healthy controls, taking into account the individual variability of the response observed in controls.
In contrast, our patient lacking NK cells failed to produce detectable IFN-γ in response to BCG alone, but produced 1,100 pg/ml/106 PBMC IFN-γ after activation with BCG plus IL-12. SCID patients that lacked both T and B cells (n=4) displayed no detectable IFN-γ production after BCG activation and a low level of IFN-γ production in response to BCG plus IL-12 (mean 4,000 pg/ml/106 PBMC). Strikingly, patients lacking both T and NK cells (n=3) produced no detectable IFN-γ in response to BCG, and very little IFN-γ in response to BCG plus IL-12 (mean 99 pg/ml/106 PBMC). In the absence of T and NK cells, these small amounts of IFN-γ were probably produced by the patients’ monocytes, detected by this in vitro blood assay. Our patient with reticular dysgenesis was unable to produce IFN-γ, even after stimulation with BCG plus IL-12 (Fig. 1C). This suggests that NK and T cells are primarily responsible for the production of IFN-γ in the blood in response to live BCG. Further investigations of a larger number of patients with NK deficiency are required to determine more accurately the relative contributions of NK and T cells, which seem to be equivalent, based on the present study.
2.3 Response of patients with immune defects impairing T cell/antigen-presenting cell cooperation (group 2)
We analyzed, in group 2, six primary immunodeficiency diseases (Table 1) 9, 13. Patients with complete CGD (n=3), CD40 ligand (CD40L; n=2), and HLA-II (n=4) deficiency displayed normal induction of IL-12p40 and IL-12p70 (Fig. 2A, B), despite a high background production of IL-12p40 in some patients. Low, but detectable, levels of IL-12p40 associated with low to normal levels of IL-12p70 were obtained for HIES (n=2) and IRAK-4-deficient (n=3) patients, following activation with BCG or BCG plus IFN-γ. Patients with X-linked anhidrotic ectodermal dysplasia with immunodeficiency (XL-EDA-ID; n=2) also showed no or only a small increase in IL-12p40 production after activation with BCG. The levels of this cytokine increased by a factor of only 2 to 5 after BCG plus IFN-γ activation. The defect in IL-12 production in these patients was confirmed by no IL-12p70 detected in supernatants after activation with BCG, associated to low levels detected after BCG plus IFN-γ (Fig. 2A, B). Overall, these data indicate that the respiratory burst and CD40/CD40L interaction are not involved in the production of IL-12p40 and IL-12p70 in vitro after activation by BCG or BCG plus IFN-γ , whereas NEMO and IRAK-4, important triggers of NF-κB activation, play an important role in IL-12 production in humans.
Consistent with our findings of normal levels of IL-12p40 and IL-12p70 production, patients with complete CGD (n=3), HIES (n=2) and CD40L deficiency (n=2) produced amounts of IFN-γ in response to live BCG and BCG plus IL-12 similar to those produced by the controls. The three patients with IRAK-4 deficiency displayed normal responses to BCG alone, but poor responses to the addition of IL-12 to BCG. Patients with HLA-II deficiency (n=4) produced little IFN-γ after activation with BCG, and IFN-γ production levels did not normalize following the addition of IL-12. This most likely resulted from the CD4 lymphopenia observed in HLA-II deficiency 14. The patients with XL-EDA-ID (n=2) also displayed a profound defect in IFN-γ production after BCG activation, and levels of this cytokine increased little following stimulation with IL-12 plus BCG (Fig. 2C). These data indicate that the NF-κB signaling pathway plays a major role in IFN-γ production by blood cells in vitro in response to infection with live BCG. The IRAK-4-deficient patients also displayed an impaired response to BCG plus IL-12 activation in vitro.
2.4 Response of patients with specific molecular defects resulting in impairment of the IL-12/IFN-γ axis (group 3)
IL-12p40 was quantified in patients with MSMD (MIM209950, 7) (n=60). Thirty-three patients with complete IL-12Rβ1 deficiency were analyzed. Basal levels of IL-12p40 production, in the absence of stimulation, and levels of this cytokine after stimulation with BCG or BCG plus IFN-γ were similar to those in healthy controls. We also generally observed an IL-12p70 response to live BCG plus IFN-γ in these patients that was similar to that in healthy subjects.
In contrast, no IL-12p40 or IL-12p70 was detected in the blood of patients with complete IL-12p40 deficiency (n=3), regardless of the type of stimulation. Patients with complete IFN-γR1 (n=5) or IFN-γR2 (n=3) deficiency displayed normal levels of IL-12p40 production following activation with BCG, but no further response was observed following the addition of IFN-γ to live BCG (similar levels or doubling at most). No IL-12p70 production in response to BCG or BCG plus IFN-γ was detected in patients with complete IFN-γR deficiency, confirming previous reports of a complete lack of response to IFN-γ. Patients with partial IFN-γR1 deficiency (n=10) or partial STAT-1 deficiency (n=6) displayed normal IL-12p40 production in response to BCG alone, but only a weak response to the addition of IFN-γ (increase by a factor of 1.5). Neither the patients with IL-12p40 deficiency (n=3) nor those with partial or complete defects in the IFN-γ pathway (n=24) produced detectable amounts of IL-12p70 in response to BCG plus IFN-γ (Fig. 3A, B). These data confirm that IL-12p70 production by blood monocytes in response to BCG plus IFN-γ is principally controlled by the IFN-γR and the associated transcription factor STAT-1.
IFN-γ was quantified in whole blood in the same cohort of patients. Patients with complete IL-12Rβ1 deficiency produced only small amounts of IFN-γ with BCG, and displayed a complete lack of response to IL-12 (no increase of the IFN-γ production following the addition of IL-12 to BCG). The three patients with the IL12B null mutation displayed no detectable IFN-γ production with BCG. Low levels of IFN-γ production were, however, detected following activation with BCG plus IL-12, probably reflecting the response of blood cells to the exogenous IL-12 added to the medium. Patients with complete IFN-γR deficiency (n=8), partial IFN-γR1 deficiency (n=10), or partial STAT-1 deficiency (n=6) produced only small amounts of IFN-γ after activation with BCG, but displayed normal increase in IFN-γ production following the addition of IL-12 to live BCG for stimulation (Fig. 3C). Thus, the production of IFN-γ by whole blood stimulated with BCG or BCG plus IL-12 strongly depends on the IL-12 pathway, and involves both IL-12p70 and the IL-12R.
Few studies of the IL-12/IFN-γ axis have been carried out in humans with PBMC or whole blood activated by live mycobacteria 15, 16. A study reported IFN-γ production in response to stimulation with live BCG in four volunteers, completed by an in vitro whole-blood assay 15. Several studies of the IL-12/IFN-γ axis have reported the activation of PBMC or whole blood by heat-killed mycobacteria or mycobacterial antigens such as PPD, ESAT6, and CFP10 (17 and references therein). These studies aimed at describing the immune response to M. tuberculosis and developing diagnostic assays for tuberculosis. In a different perspective, Levin et al. 18, 19 reported a decrease in the level of TNF-α produced by PBMC in response to endotoxin plus IFN-γ/endotoxin and in levels of IFN-γ in response to mycobacterial antigens in IFN-γR1 deficiency. Holland et al. 20 later reported a 10% decrease in IL-12p40 and IFN-γ production in response to PHA in the PBMC of two patients with MSMD due to loss-of-function mutations in IFN-γR1. They also reported a decrease in PHA-induced IFN-γ production in a patient with a mutation affecting the extracellular domain of IFN-γR2 21.
However, the cellular basis of IL-12 and IFN-γ production, as well as that of the response to IL-12 and IFN-γ, upon blood stimulation by live or even dead mycobacteria, has not been determined. Whole blood cultures and stimulation by live mycobacteria enable the evaluation of the contributions and reciprocal interactions of all cell types and molecules in the sample. Our study is the first to investigate a large cohort of patients (n=90) with such a variety (n=18) of specific inherited immunodeficiencies to dissect the IL-12/IFN-γ axis at the cellular and molecular level. Our study mostly dealt with small groups of patients and we cannot exclude the possibility that inter-individual variability would somewhat change the global picture.
The normal IL-12p70 production in most patients lacking T, B, NK, or PMN cells and the lack of IL-12p70 production in the patient with reticular dysgenesis, suggest that the major blood cells responsible for IL-12p70 production in response to BCG and BCG plus IFN-γ are probably monocytes (including bona fide monocytes and dendritic cells), although the absence of a specific human monocyte defect or defect of monocytes/dendritic cells precludes definitive conclusions 22. IL-12p70 production is also strongly dependent on the NF-κB pathway, as demonstrated by the diminished IL-12 production in patients with NEMO and IRAK-4 deficiency. Our results are also consistent with IL-12 production in response to mycobacteria and IFN-γ being largely independent of molecules such as CD40L, HLA-II and of the respiratory burst. However, this IL-12 response to live BCG is controlled by the IL12B gene and IL-12 production in response to BCG plus IFN-γ is heavily dependent on the presence of functional IFN-γR1, IFN-γR2, and STAT-1 molecules.
NK and T cells have been shown to make a major contribution to IFN-γ production in response to BCG. Patients lacking NK or T cells or both NK and T cells displayed similar profound defects in IFN-γ production following stimulation with BCG alone or BCG plus IL-12 (less pronounced than patients lacking both NK and T cells). In contrast, neither B cells nor PMN cells seem to be involved in the IFN-γ production, as demonstrated by the normal levels of IFN-γ production in this assay for patients with Kostmann's and Bruton's diseases. We were also able to suggest the importance of molecules such as IRAK-4 and NEMO which contribute to IFN-γ production, induced by IL-12 activation of NK and T cells. Similarly, the absence of other molecules, such as CD40L and components of the gpPHOX complex, had no detectable effect on the IFN-γ production induced by live BCG. We confirmed with MSMD patients that IFN-γ production in response to BCG infection depends on IL-12/23 priming, and that IFN-γ production in response to BCG plus IL-12 heavily depends on the IL-12/23 pathway, particularly on the integrity of the IL-12Rβ1 molecule.
This study has also significant clinical implications, as this assay can be used to identify deficient pathways in patients with high levels of susceptibility to mycobacteria, and could therefore be used to search directly for mutations. This test proved to be particularly useful for the screening of patients with MSMD or a suspicion of NEMO or IRAK-4 mutation. The known genetic etiologies of MSMD (complete IFN-γR1 or partial IFN-γR1 and complete IFN-γR2 deficiencies, partial STAT-1 deficiency, complete IL-12p40 deficiency, and complete IL-12Rβ1 deficiency) were successfully diagnosed at the molecular level, following an initial screening with our whole-blood assay. Furthermore, the rapid diagnosis of complete IFN-γR1/2 deficiencies in infected patients was confirmed by the detection of IFN-γ in the patient's serum 23. Our blood assay appears to be specific and sensitive to successfully identify impaired pathways in the IL-12/IFN-γ circuit and guide the search for disease-causing genes in patients with MSMD.
4 Materials and methods
4.1 Subjects and patients
We compared three different groups of patients with adult local (n=50) healthy subjects. Mean age (standard deviation) was 34 years (6.5) for controls and 10.5 years (9) for patients. For group description see Table 1. Our study was conducted according to the principles expressed in the Helsinki Declaration, with informed consent obtained from each patient or the patient's family. The genetic defects were identified in all patients from group 3 and in some, but not all, patients from groups 1 and 2. The diagnosis criteria were clinical and immunological, following current states of knowledge 10.
Group 1 included 14 patients lacking a specific blood cell type. For description see Table 1. The SCID patients do not have detectable autologous T cells in the blood. Group 2 included 14 patients with primary immunodeficiency diseases other than MSMD (for description see Table 1). All patients with complete CGD had no detectable respiratory burst. Group 3 included 60 patients with MSMD due to recently identified molecular defects 6.
4.2 Whole-blood cultures and activation by live BCG
Venous blood samples were collected into heparinized tubes. They were diluted 1:2 in RPMI 1640 (GibcoBRL) supplemented with 100 U/ml penicillin and 100 µg/ml streptomycin (GibcoBRL). We dispensed 6 ml of the diluted blood sample into 4 wells (1.5 ml/well) of a 24-well plate (Nunc). It was then incubated in a two-stage procedure during 18 and 48 h at 37°C in an atmosphere containing 5% CO2/95% air, and under four different conditions of activation: with medium alone, with live BCG (M. bovis BCG, Pasteur® sub-strain) at an MOI of 20 BCG/leukocytes, with BCG plus IFN-γ (5,000 IU/ml; Imukin®, Boehringer Ingelheim) and with BCG plus recombinant IL-12p70 (20 ng/ml; R&D Systems®). An MOI of at least 20 in individuals without any cytopenia was used. The first incubation stage was completed after 18 h of culture, 450 µl supernatant was collected from each culture well and frozen at –80°C. After 48 h, by the end of the second incubation stage, whole remaining volume of each well was recovered, centrifuged at 1,800×g for 10 min, and the supernatant was stored frozen at –80°C until analysis. For patients whose blood samples were transported from elsewhere, we also analyzed a “travel” control in parallel, when available.
4.3 Cytokines ELISA
Cytokine concentrations were analyzed by ELISA, using the human Quantikine IL-12p70 HS and IL-12p40 kits from R&D Systems and the human Pelikin™ or Pelipair IFN-γ kit from Sanquin, according to the manufacturers’ guidelines. These kits were applied using matched antibody pairs. Optical density was determined using an automated MR5000 ELISA reader (Thermolab Systems).
Quantitative analysis was carried out using the non-linear four-parameter logistic (4PL) calibration model developed by O'Connell 24. An in-house software based on Microsoft Excel® application language was developed for this purpose. Intermediate results for each cytokine are expressed in pg/ml. However, PBMC counts vary according to the subject, and are dependent on age, in particular. We therefore standardized the final results by expressing them per million PBMC, in the unit pg/ml/106 PBMC. The number of PBMC was determined from blood cell counts carried out on day 0.
4.4 Statistical analysis of the data
An initial Q-plot statistical study demonstrated that cytokine data were not normally distributed for the healthy population (controls). These data were log-transformed, and the resulting distribution generally approximated a normal distribution.
The effect of gender and age on IL-12p40, IL-12p70, and IFN-γ levels under four different sets of activation conditions (no stimulation, stimulation with BCG alone, stimulation with BCG plus IFN-γ, and stimulation with BCG plus IL-12) was assessed by the means of one-way analysis of variance for gender and linear regression analysis for age. Intra-individual correlation of IL-12p40, IL-12p70, and IFN-γ values was taken into account for these analyses. All computations were made with the generalized linear model (GLM) procedure of SAS software v8.2 (SAS Institute, Cary, NC).
We would like to thank Laurent Abel and members of the laboratory of HGID as well as Stéphane Blanche, Alain Fischer and members of the Pediatric Immunology Hematology Unit for helpful discussions, Claude Frehel for advice in BCG culture and titration, Françoise Le Deist for the diagnosis of SCID, HLA-II- and CD40L-deficient patients and helpful advices, Marie-Anne Pocidalo for the diagnosis of CGD patients, and Jean Donnadieu and Françoise Valensi for the diagnosis of patients with neutropenia. We also thank all internits and pediatricians world-wide who have kindly accepted to send us blood samples from their patients, whom we also thank warmly together with their families for their participation.