Abstract
We here present the first genetic fine mapping of experimental autoimmune neuritis (EAN), the animal model of Guillain-Barré syndrome, in a rat advanced intercross line. We identified and refined a total of five quantitative trait loci on rat chromosomes 4, 10, and 12 (RNO4, RNO10, RNO12), showing linkage to splenic IFN-γ secretion and disease severity. All quantitative trait loci were shared with other models of complex inflammatory diseases. The quantitative trait locus showing strongest linkage to clinical disease was Ean6 and spans 4.3 Mb on RNO12, harboring the neutrophil cytosolic factor 1 (Ncf1) among other genes. Polymorphisms in Ncf1, a member of the NADPH oxidase complex, have been associated with disease regulation in experimental arthritis and encephalomyelitis. We therefore tested the Ncf1 pathway by treating rats with a NADPH oxidase complex activator and ameliorated EAN compared the oil-treated control group. By proving the therapeutic effect of stimulating the NADPH oxidase complex, our data strongly suggest the first identification of a gene regulating peripheral nervous system inflammation. Taken together with previous reports, our findings suggest a general role of Ncf1 and oxidative burst in pathogenesis of experimental autoimmune animal models.
Guillain-Barré syndrome (GBS)4 is an inflammatory disease of the peripheral nervous system consisting of different subtypes with acute inflammatory demyelinating polyradiculoneuropathy (AIDP) being the most prevalent subtype (1). GBS is clinically defined as a peripheral neuropathy causing limb weakness that progresses for up to 4 wk (2). Several clinical observations and epidemiological studies indicate a correlation between GBS and preceding infections, with Campylobacter jejuni being the most comprehensively linked pathogen, suggesting an infectious etiology at least for a subset of GBS cases (3). It has also been shown that activated T cells and macrophages, as well as Abs against components of peripheral nerve myelin (PNM), are important for the pathogenesis of GBS (4, 5, 6). Although advances have been made in understanding the etiopathogenesis, the inflammatory disease mechanisms regulating GBS remain elusive. Accordingly, the accepted treatments of GBS, that is, plasma exchange (7) and i.v. application of immune globulins (8), are not targeted toward a specific disease mechanism.
Siblings of GBS-affected individuals have an increased risk of developing GBS compared with the general population, suggesting a genetic component in GBS (9). Association between different HLA types and GBS has been reported with preceding C. jejuni infection but has only been observed in some populations (9, 10, 11, 12). Polymorphisms in the TNF gene have been associated with a rare GBS subtype in a Chinese population (13), but association studies have so far failed in identifying single genes regulating the more prevalent AIDP subtype. This indicates a genetically complex regulation of GBS, and large homogeneous patient cohorts are required for accurate genetic analysis. The relatively low number of patients makes it difficult to employ genome-wide approaches to identify susceptibility genes. An alternative approach to gain better understanding of the immunopathogenesis of GBS is through use of the animal model experimental autoimmune neuritis (EAN). EAN can be induced with bovine PNM (bPMN) and is clinically characterized by ascending peripheral paresis. It closely resembles the pathology of the AIDP subtype with CD4+ T cell-mediated responses against the peripheral myelin sheath (7). The benefits of animal models include reduction of genetic heterogeneity through use of inbred strains, well-defined and accepted disease phenotypes, large numbers of animals, and controllable environmental factors.
A previous linkage study in an MHC-identical (DA (RT1av1) × ACI (RT1av1))F2 rat population showed that EAN is regulated by multiple quantitative trait loci (QTLs) (14). A second study demonstrated MHC as a major susceptibility factor for experimental autoimmune neuritis as PVG (RT1c) rats are resistant to bPNM immunization while PVG rats with the RT1av1 haplotype develop an acute form of EAN (15). Furthermore, a significant overlap has been detected between QTLs that regulate EAN and other experimental inflammatory diseases, such as experimental autoimmune encephalomyelitis (EAE) and experimental arthritis (EA) (14), suggesting shared genes and pathogenic mechanisms.
We herein present the first genetic fine mapping of EAN using an advanced intercross line (AIL). An AIL is created by random intercrossing of two different inbred animal strains for several generations, resulting in more recombinations between two loci (16). As the MHC region has already been described to regulate EAN (15), hereby called Ean1, we chose to cross the MHC-identical susceptible DA rats (RT1av1) with PVG-RT1a rats (RT1av1), which carries the resistant PVG background. Our interest here was two-fold: 1) to fine-map the non-MHC regulatory QTLs revealed in the F2 analysis, and 2) to map other established QTLs from the related inflammatory disease models EAE and pristine-induced arthritis (PIA).
We identified and refined five EAN-regulating QTLs in our AIL. One of these loci, Ean6 on rat chromosome 12 (RNO12), has been described to regulate disease outcome both in EAE (Eae5) and PIA (Pia4). Fine mapping of Pia4 in a rat AIL positionally cloned the neutrophil cytosolic factor 1 (Ncf1) as the responsible arthritis severity regulating gene (17). Moreover, investigation of Eae5, which controls disease severity in EAE (18), disclosed Ncf1 within a narrow confidence interval by using a G7 advanced intercross line (19). In view of these earlier findings, we sought to investigate the role of Ncf1 as the disease-regulating gene of Ean6. We treated animals with an agonist of the NADPH oxidase complex for which Ncf1 encodes a protein. Rats treated in a prophylactic setting showed a dramatic reduction in EAN disease score compared with oil-treated controls.
Materials and Methods
Rats and advanced intercross line
The AIL originated from MHC-identical DA and PVG-RT1a rats. DA rats were originally obtained from the Zentralinstitut für Versuchstierzucht (Hannover, Germany), and PVG-RT1a rats were originally obtained from Harlan U.K. The F1 generation was established by breeding two pairs of DA and PVG-RT1a, as female founders. The F2 generation was produced from seven couples each of F1 rats with DA and PVG-RT1a female founders, respectively. Fifty F2 breeding pairs with both DA and PVG-RT1a female founders generated the third generation. All following generations were created by random breeding of 50 males and females, avoiding brother-sister mating. Animals were bred and housed at the Karolinska University Hospital (Stockholm, Sweden) or at Scanbur (Sollentuna, Sweden) in polystyrene cages containing aspen-wood shavings and with free access to food and standard rodent chow with a 12 h light/dark cycle. The animals were routinely monitored for pathogens according to a health-monitoring program for rats at the National Veterinary Institute (Uppsala, Sweden). Four hundred fifty rats (215 males and 235 females) from the 12th generation where shipped to Tübingen, Germany, for the G12 experiment.
Induction and clinical evaluation of EAN
PNM was prepared from a bovine sciatic nerve as described previously (20). In both experiments, rats were between 8 and 14 wk old at the time of immunization. Rats were anesthetized by ether inhalation and injected intradermally at the base of the tail with a total volume of 200 μl of inoculum containing 500 μg of bPNM in 100 μl of PBS and 100 μl of IFA. For the G12 experiment, rats were immunized in two sessions on successive days. Rats were monitored daily for clinical signs of EAN and weighed every second day up to days 34–37 postimmunization (p.i.) for the G12 experiment, and up to day 42 in the treatment study. Clinical signs of EAN were graded as follows: 0, normal; 1, less active, reduced tone of tail; 2, limp tail, impaired righting; 3, absent righting; 4, gait ataxia; 5, mild paraparesis; 6, moderate paraparesis; 7, severe paraparesis; 8, tetraparesis; 9, moribund; 10, death (21). Weight loss was calculated as the difference between weight on day 7 p.i. and lowest weight until the end of experiments. Maximum disease score was defined for each rat as the highest observed clinical score during the experiment. Cumulative disease score was defined as the total sum of each day’s clinical score. Weight loss is defined as the relative difference between the lowest weight during the experiment and the weight at day 0 (start of experiment).
Phytol experiments
We divided animals into two treatment groups, one early stage and one late stage. Seven or eight DA female rats per group were treated at day 0 with 100 μl of the phytol side chain of vitamin E (α-tocopherol), an activator of the NADPH oxidase complex. Seven or eight control rats per group were given 100 μl of olive oil. All rats were injected s.c. at the tail base and were immunized with bPNM as described above at the day of treatment (day 0).
Anti-bPNM determination
In the genotyping study as well as in the treatment study, sera were sampled from the heart of each rat after sacrifice. ELISA plates (96-well; Nunc) were coated with 100 μl/well protein (bPNM) dilution at a concentration of 2.5 μg/ml overnight at 4°C. Plates were washed with PBS/0.05% Tween and blocked with 5% milk powder for 1 h at room temperature. After washing, diluted serum samples were added and incubated for 1 h at room temperature. After washing, rabbit anti-rat antiserum was added and plates were incubated for 1 h at room temperature. After another washing step peroxidase-conjugated goat anti-rabbit antiserum was added. After incubation of 30 min, plates were washed. Bound Abs were made visible by addition of ABTS. OD was read at 405 nm (22).
ELISPOT for IFN-γ production
The number of IFN-γ-secreting mononuclear cells was measured with the ELISPOT method using a standard protocol (22). A total of 111 rats (52 males and 59 females) that displayed varying disease severity were randomly selected. Mononuclear cells (MNCs) were isolated from spleens taken between day 34 and day 37 for both the G12 experiment and the late stage of the phytol experiment. For the early stage of the phytol experiment, draining lymph nodes were taken on day 11. After incubation for 48 h the cells were discarded and the plates were washed with PBS. Rat IFN-γ bound to DB1 were visualized with biotinylated mouse mAb DB12, avidin-biotin peroxidase, and carbazole. Isolated MNCs were both assayed directly upon isolation and recalled with peripheral nervous system Ags. Ags were applied in triplicates at a final concentration of 10 μg/ml. The Ags applied were myelin oligodendrocyte protein (MOG) peptide containing aa 73–90 as a control Ag, protein P2 peptide containing aa 58–81, whole bPNM, Con A, and no Ag.
Genotype analysis
Nine genomic regions on RNO2, 3, 4, 6, 10, 12, 15, 17, and 18 were covered in the genotyping of the G12 experiment. These represent all genomic regions displaying p values <0.01 from the previous EAN F2 intercross (14), in total comprising seven loci. Furthermore, nonoverlapping known QTLs from the related EAE and EA models identified in the relevant strain combinations were also tested, resulting in the addition of two more loci and a total of nine tested regions. Genomic DNA was prepared from tail tips as described previously (23). Polymorphic microsatellite markers were used for genotyping. DNA amplification was performed with PCR, and the forward primers were either end-labeled with [γ-33P]ATP or with a fluorescent dye (VIC, NED, FAM, PET) (24). Primers were obtained from Proligo or Applied Biosystems. [γ-33P]ATP-labeled PCR products were size-fractionated on 6% polyacrylamid gel and visualized by autoradiography. Fluorescently labeled products were run on ABI 3730 capillary sequencer and analyzed with the software GeneMapper v3.7 (Applied Biosystems) (25). All genotypes were evaluated manually and double-checked. DNA was collected and genotyped from 450 rats, for which we had 437 correlating phenotypes due to the fact that 13 animals died a short time after immunization without having developed EAN signs.
Statistical analysis
Linkage analysis was performed with the statistical software R/qtl version 2.6.1. Clinical disease phenotypes were tested for linkage using the Haley-Knott regression model except for disease incidence that was tested with a binary model, as the phenotype outcome is binary. The number of IFN-γ-secreting splenocytes was tested for linkage using the imputation model, as only 111 of 437 animals were assayed. Differences in allelic effects of QTLs were analyzed with the nonparametric Mann-Whitney (for two groups) and Kruskal-Wallis tests (for three groups) using Graphpad Prism 5.0 (GraphPad Software). The 95% confidence intervals (CI) for QTLs were arbitrarily defined by first determining a drop in a 10-base logarithm of the likelihood ratio (LOD) of 1.8, based on simulation experiments performed previously (26). We then used the flanking markers for CI definition. All shown p values were calculated using the Mann-Whitney U test in Graphpad Prism 5.0.
Results
Disease characteristics of G12 (DA × PVG-RT1a) rats
The incidence of EAN in the 437 rats used in the G12 experiment was 83.5%: 94.4% in female rats and 71.2% in male rats. No significant differences were observed in disease onset between the sexes. Males, however, showed a less severe disease course with a lower maximum score compared with female rats, as well as a lower cumulative score (Table I⇓). We also observed that changes in weight correlated with disease severity (Fig. 1⇓).
Weight loss correlates with EAN severity. Correlation between weight loss and EAN cumulative score in the (DA × PVG-RT1a) AIL experiment. The curve was fitted with linear regression using GraphPad Prism 5.0.
Disease characteristics of G12 (DA × PVG-RT1a) AIL experiment
Abs and cell-mediated immune responses in EAN AIL G12 rats
We investigated adaptive immune responses by assaying both humoral and cellular activities in relevant tissues. Anti-bPNM-IgG Ab levels were determined in the (DA × PVG-RT1a) AIL rats using ELISA assays. Serum samples were available from 433 of the 437 rats. No significant differences were observed in anti-bPNM-IgG Ab titers between sexes or in clinical severity (data not shown). Enumeration of cells secreting IFN-γ was determined by the ELISPOT assay. We observed a mild correlation between IFN-γ secretion and cumulative disease score (Fig. 2⇓). No significant differences in IFN-γ secretion were observed between the selected males and females (supplemental Table I).5
Correlation between IFN-γ secretion and EAN severity. Correlation between IFN-γ secretion of MNC from spleen collected day 35 and cumulative EAN score in the (DA × PVG-RT1a) AIL experiment. Curve was fitted with linear regression using GraphPad Prism 5.0.
Linkage analysis of the G12 experiment
Linkage analysis in the G12 population was performed on nine genomic regions previously identified in F2 crosses of several inflammatory diseases. We identified a total of five QTLs regulating susceptibility to EAN or immunological phenotypes relevant for peripheral autoimmune neuroinflammation, hereby called Ean2–6. The QTLs confirm previously suggested loci (Fig. 3⇓). For Ean2 on RNO4 we could observe linkage to a number of IFN-γ-secreting splenocytes, obtaining a LOD score of 6.4 (Fig. 4⇓A and Table II⇓). On RNO10 we could identify three distinct QTLs, hereby called Ean3-5. Ean3 reaches a LOD score of 3.1 for suggestive linkage with weight loss. Ean4 as well as Ean5 were linked to the number of IFN-γ secreting splenocytes (Fig. 4⇓B and Table II⇓). For Ean6, the highest linkage was seen to disease severity (Fig. 4⇓C and Table II⇓) and was thus the only region showing linkage to a clinical phenotype. The CI for Ean6 spans 4.3 Mb and harbors 32 confirmed genes and 11 predicted genes with orthologs in other species (Tables II⇓ and III⇓, adapted from www.ensembl.org). Ncf1, which lies within the CI, has previously been identified to regulate both EAE and PIA, making it an interesting candidate also in EAN. Alleles from the PVG strain confer the increase in weight loss (Ean3), whereas alleles from the DA strain confer the higher IFN-γ secretion (Ean2, 4, 5) and more severe disease (Ean6) (Fig. 5⇓). Ean2 and Ean3 display similar effects in both sexes, whereas Ean4–6 are only significant in females, in a stratified analysis (supplemental Fig. 1).
Ean2–6 with QTLs in EAE and EA. Schematic overview of EAN QTLs on rat chromosome 4, 10, and 12 mapped in G12 with their shared QTLs from mapping in intercrosses of EAE and EA. Black lines depict the genomic regions covered in G12. The peak markers from the previous EAN F2 cross are shown on the left-hand side.
The AIL linkage analysis identifies five QTLs. QTLs and LOD scores on RNO4 (A), RNO10 (B), and RNO12 (C) in the (DA × PVG-RT1a) AIL experiment. The y-axis depicts the log likelihood for the tested phenotype and the x-axis shows the microsatellite markers used and their genomic positions. For Ean2, linkage to IFN-γ secretion was observed with a LOD score of 6.4. Ean3 linked to weight loss with a LOD score of 3.1 (gray dotted line). Ean4 and Ean5 linked to IFN-γ secretion with LOD scores of 4.0 and 4.7, respectively (black dotted lines). Ean6 linked to maximum disease score with a LOD score of 4.9 (black line) and to cumulative disease score with a LOD score of 4.1 (gray line).
Allelic effects of Ean2–6. Allelic effects of the peak markers from the G12 experiment. DA/DA and PVG/PVG represent animals homozygous for DA or PVG alleles at the marker, respectively. DA/PVG represents heterozygous animals. DA alleles within Ean2, 4, 5, 6 (A and C–E) confer increase in disease phenotypes, whereas PVG alleles confer the increased weight loss in Ean3 (B).
Summary of identified EAN regulating QTLs
Candidate genes within 95% CI of Ean6 at RNO12
Treatment with an NADPH oxidase complex activator ameliorates EAN
To test the importance of the Ncf1 pathway in EAN, we performed a prophylactic treatment of bPNM-immunized DA females with phytol on day 0, which led to a reduction in disease incidence (Fig. 6⇓A). Incidence of EAN was 37.5% in the phytol-treated group and 87.5% in the oil-treated control group. Additionally, the phytol-treated group displayed a less severe disease compared with the oil-treated group, whereas disease onset was identical for both groups (p < 0.01) (Fig. 6⇓B and Table IV⇓).
Phytol ameliorates EAN. DA rats were treated with 100 μl of phytol or olive oil as control s.c. in the base of the tail on day 0 (phytol, n = 8; olive oil, n = 8). A, EAN disease score until day 42 p.i. B, Cumulative score after 42 days p.i. Bars indicate the mean ± SD.
Disease characteristics after prophylactic treatment with phytol
Investigation of adaptive immune response cells upon phytol treatment
To evaluate the effects of phytol treatment on adaptive immune responses, we first studied autoantibody levels by measuring anti-bPNM Ab titers with ELISA from sera collected on days 11 and 42 p.i. No differences in anti-bPNM-IgG Ab titers were observed between phytol- and oil-treated animals at either of the time points (data not shown). Second, inflammatory effects mediated by mononuclear cells were evaluated using ELISPOT to measure the number of IFN-γ-secreting cells upon stimulation from draining lymph nodes on day 11 p.i. and from splenic mononuclear cells at day 42 p.i. At an early disease stage we observed no differences in the number of IFN-γ-secreting splenocytes between phytol- and oil-treated animals regardless of the Ag used. In contrast, for late-stage disease, phytol-treated rats showed a clear reduction in IFN-γ-secreting lymph node cells compared with oil-treated rats (Fig. 7⇓).
IFN-γ-secreting lymph node cells in rats treated with phytol. A, Draining lymph nodes were collected on day 11 p.i. and were assayed for IFN-γ production with ELISPOT and (B) on day 42 p.i. in rats treated with either phytol or olive oil (control group). There were seven rats for each group in day 11 collection and eight rats for each group in day 42 collected lymph nodes. Bars indicate the mean ± SD.
Discussion
This study presents the first high-resolution linkage analysis for EAN, using a 12th generation AIL. We identified five QTLs overlapping already described loci for the chronic inflammatory and thus EAN-related models EAE and EA (17, 19, 27, 28, 29). This indicates a number of shared mechanisms between these complex inflammatory disease models converging at common inflammatory key events, despite their differences in etiology.
Ean6 provided the strongest linkage to clinical severity phenotypes such as maximum and cumulative disease score. Disease onset, however, could not be linked to Ean6, likely due to the high disease incidence. Ean6 overlaps Eae5 and Pia4, whose effect has been positioned to Ncf1, shown to regulate severity in PIA as well as encephalomyelitis in both mice and rats (17, 30). It is possible, however, that the EAE-implicated CLDN4 or additional genes within the Ean6 region contribute to EAN disease severity (19). Importantly, other genes involved in adaptive immune responses exist within the Ean6 interval, such as linker for activation of T cells family member 2 (LATFM2) that encodes a protein on B cells, NK cells, mast cells, and mononuclear cells and promotes activation of these cells (31).
Ncf1 encodes a protein of the NADPH oxidase complex, and differences in oxidative burst correlate with disease susceptibility among different rodent strains. As this suggested Ncf1/oxidative burst as the primary candidate gene/mechanism in Ean6, we treated DA rats with the phytol side chain of vitamin E, a known activator of the NADPH oxidase complex. Vitamin E itself acts as an antioxidant vitamin, whereas the effect of vitamin E-dependent NADPH oxidase activation results in an elevated oxidative burst (17). Phytol treatment led to strong reduction in EAN disease severity and a lower number of IFN-γ-secreting cells in late disease stage. We also investigated IFN-γ secretion at an earlier time point, day 11 p.i., to see how early phytol can mediate a reduced IFN-γ secretion, as this would indicate a role in disease priming or during disease effector mechanisms. We studied MNCs from draining lymph nodes, generally considered a good source of effector cells in early disease in similar animal models. However, we observed no differences in IFN-γ secretion in draining lymph nodes collected before disease onset. One possibility is that the cellular population accounting for the immunoregulating cell population has already left the lymph nodes into the blood stream. A kinetic study of draining lymph nodes from MOG-immunized DA and PVG.AV1 rats collected at various time points shows that there is a significant reduction of proliferating lymph node cells at day 12 p.i. compared with day 7 p.i. (Mélanie Thessén Hedreul and Alan Gillett, unpublished data).
We here propose that naturally occurring polymorphisms in rodent strains within Ncf1 or within Ean6 regulating Ncf1 activity regulate levels of oxidative burst and EAN severity. Oxidative bursts from macrophages reduce T cell responsiveness in experimental autoimmune models (32), and in humans high levels of reactive oxygen species have been implicated in the pathogenesis of different autoimmune diseases, including GBS (33). Accordingly, one recent study showed that isolated peripheral blood leukocytes from severely affected GBS patients, upon NADPH oxidase complex stimulation, have lower production of oxygen radicals as compared with leukocytes of patients with less severe GBS (34). This interesting finding pinpoints at least some of the beneficial roles of the oxidative burst to the NADPH oxidase complex, and it is in agreement with our proposed correlation between Ncf1/NADPH oxidase complex stimulation, mononuclear cell activity (IFN-γ secretion), and ultimately EAN severity.
Three of our described QTLs, Ean2, Ean4, and Ean5, are primarily linked to a number of IFN-γ-secreting cells, a phenotype we show can correlate with disease severity. This phenotype reflects subclinical differences in the number of reactive immune cells in lymphoid tissues. It is likely that the IFN-γ production is largely attributable to T cells. In splenocytes the two major producers of IFN-γ are T cells and NK cells. T cells have a more sustained and delayed production of IFN-γ, which would better explain our observed differences in late-stage disease. However, we cannot exclude effects from NK cells or other cell types. We here argue for the use of nonclinical immunological phenotypes that correlate with clinical disease, for example, the number of IFN-γ-secreting mononuclear cells. As these phenotypes are less complex than clinical disease outcome, they can pick up linkage with regions modulating the pathogenesis before clinical symptoms. This is particularly useful in mild disease induction with relatively low clinical disease severity.
As an additional phenotype, we measured total IgG levels against bovine peripheral nerve myelin components but could not reveal any differences between the different affliction states in late disease. This is congruent with the results we have seen in our treatment studies. Additionally, genetic regions tested showed no linkage to anti-bPNM Abs in the G12 experiment, although it is possible that the phenotype is regulated by other genomic regions not mapped here. Our findings are in line with previous studies showing that B cells do not sustain EAN and that Ab transfer alone does not induce EAN but also that CD4+ cells are needed for induction (35, 36).
Ean3 showed linkage to weight loss during disease. In EAE, weight loss correlates with disease progression (37) and is often regulated by disease severity and susceptibility loci. Interestingly, some loci regulating EAE and other models are linked primarily to weight loss (38, 39), indicating the presence of a distinct disease regulation at least partially independent of adaptive cell responses. Nevertheless, the mechanisms governing the weight loss in EAN remain unclear.
The overall incidence in the AIL experiment was higher in females compared with males, which is relevant when studying gender differences in Ean2–6. For IFN-γ secretion, the loci Ean4, Ean5, and Ean6 are only significant in females. This can be explained by the more severe disease observed in females. Ean2 and Ean3 have similar effects in both sexes even though power is lower in the stratified analysis. As IFN-γ secretion in both sexes correlates with disease, it is likely that Ean2 is a stronger regulator of the IFN-γ secretion, allowing even males to display linkage.
Our data emphasize the importance of animal models for understanding etiology and pathogenesis of complex genetic autoimmune diseases. The identified QTLs in this study, together with the previously described non-MHC regions as well as the MHC itself, argue for EAN being an autoimmune disease under a strong genetic control (14, 15). To summarize the data of the present study we have identified and confirmed five QTLs, all shared with other disease models, in a first genetic fine mapping analysis of EAN using an advanced intercross line. The QTL showing strongest linkage to clinical disease is Ean6 on RNO12, harboring Ncf1. By showing the therapeutic disease-protective effect of stimulating the NADPH oxidase complex, for which Ncf1 codes a protein, our data provide evidence of the first identification of a gene regulating peripheral neuroinflammation. Ncf1 and the NADPH oxidase complex therefore serve as potential pharmacogenetic targets for therapeutic intervention in peripheral nervous system inflammation.
Disclosures
R.W. is currently employed by Merck Seronof., Switzerland; P.O. is employed by Redoxis AB.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
↵1 This study was supported by the German Research Foundation (DFG We 1947/4-1/2) to R.W. and European Rat Tools for Functional Genomics (European Commission contract no. LSHG-CT-2005-019015) and Neuroprotective Strategies for Multiple Sclerosis (European Commission contract no. LSHM-CT-2005-01863) to T.O.
↵2 A.H. and A.D.B. contributed equally to this work.
↵3 Address correspondence and reprint requests to Dr. Robert Weissert, Department of Neurosciences, Division of Neurology, Geneva University Hospital, Micheli-du-Crest 24, 1211 Geneva 14, Switzerland. E-mail address: robert.weissert{at}hcuge.ch
↵4 Abbreviations used in this paper: GBS, Guillain-Barré syndrome; AIDP, acute inflammatory demyelinating polyradiculoneuropathy; AIL, advanced intercross line; bPMN, bovine peripheral nerve myelin; CI, confidence interval; EA, experimental arthritis; EAE, experimental autoimmune encephalomyelitis; EAN, allergic autoimmune neuritis; LOD, logarithm of the likelihood ratio; MNC, mononuclear cell; MOG, myelin oligodendrocyte glycoprotein; Ncf1, neutrophil cytosolic factor 1; p.i., postimmunization; PIA, pristine-induced arthritis; PNM, peripheral nerve myelin; QTL, quantitative trait locus; RNO, rat chromosome; RT1, MHC of rat.
↵5 The online version of this article contains supplemental material.
- Received November 17, 2008.
- Accepted January 26, 2009.
- Copyright © 2009 by The American Association of Immunologists, Inc.
References
In this issue
