If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
NIH Eastern Regional Comprehensive Metabolomics Resource Core, Department of Nutrition, University of North Carolina at Chapel Hill, Chapel Hill, NCDepartment of Nutrition, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC
NIH Eastern Regional Comprehensive Metabolomics Resource Core, Department of Nutrition, University of North Carolina at Chapel Hill, Chapel Hill, NCDepartment of Nutrition, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC
NIH Eastern Regional Comprehensive Metabolomics Resource Core, Department of Nutrition, University of North Carolina at Chapel Hill, Chapel Hill, NCDepartment of Nutrition, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC
Toxicology Branch, Division of National Toxicology Program, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC
Environmental influences play a significant role in host immune responses, and it is hypothesized that environmental changes might be responsible for enhanced allergy prevalence. The hygiene hypothesis proposed that smaller families and increased cleanliness reduce infectious diseases (and thereby endotoxin exposure) in a household and contribute to enhanced allergic disease.
National Academies of Sciences, Engineering, and Medicine Finding a path to safety in food allergy: assessment of the global burden, causes, prevention, management, and public policy.
Vaccines can reduce microbial exposure by preventing infections, and vaccine adjuvants, such as alum, can promote allergies in genetically predisposed populations.
Alum induces TH2 responses to coadministered antigens and potentially to unrelated environmental allergens, thus providing bystander (heterologous) responses that contribute to allergic disease.
Modification of childhood vaccines by the addition of TH1-enhancing Toll-like receptor ligand adjuvants, such as monophosphoryl lipid A (MPL) or CpG, to alum might reduce the TH2-skewing activity of alum and balance host immunity to reduce the development of allergic disease.
In this study we used a mouse model of peanut allergy to test the hypothesis that environmental endotoxin, alum, or both influence the severity of allergen-specific hypersensitivity. Toll-like receptor 4–responsive C3H/HeOuJ mice were housed in low (paper) or standard (corncob) endotoxin bedding for 1 month before alum-adjuvanted ovalbumin (OVA) immunization and peanut sensitization. (A detailed experimental design can be found in the Methods section and Table E1 in this article's Online Repository at www.jacionline.org.) Although endotoxin levels were significantly lower in paper bedding compared with those in corncob bedding before animal exposure, bedding endotoxin levels were similar after 1 day (see Fig E1, A, in this article's Online Repository at www.jacionline.org). Multivariate analysis of the broad-spectrum metabolomics data distinguished mice based on bedding type (see Fig E1, B), and the metabolites identified (see Table E2 in this article's Online Repository at www.jacionline.org) suggested that pathways related to amino acid metabolism and transport might be important in differentiating the type of bedding in the animal's living environment (see Table E3 in this article's Online Repository at www.jacionline.org).
OVA/alum vaccines were administered subcutaneously to provide alum exposure before gastric peanut sensitization to evaluate alum's influence on allergen hypersensitivity. Although TH1-inducing adjuvants used prophylactically or coadministered with allergens reduce the development of allergen-specific hypersensitivity reactions in mice, widespread prophylaxis with TH1-inducing adjuvants to reduce the development of allergy in human subjects is not practical. However, it might be practical to modify current vaccine formulations to contain alum and TH1-biased adjuvants to induce a balanced TH1/TH2 response and reduce the development of allergic disease. Therefore we evaluated the TH1-inducing adjuvants MPL or CpG combined with alum for their ability to enhance OVA-specific TH1 responses and influence peanut allergy outcomes.
Vaccine antigen-specific serum antibody levels were measured to evaluate adjuvant-modulated humoral immunity (Fig 1, A). OVA-immunized mice had anti-OVA IgG1 and IgG2a serum titers that were significantly greater than unvaccinated mice. Anti-OVA IgG1 levels were significantly greater in paper-housed mice vaccinated with OVA/alum/CpG than in corncob/OVA/alum or OVA/alum/CpG mice or paper/OVA/alum/MPL mice. Alum significantly increased OVA IgE responses compared with peanut alone regardless of environment and in paper-housed mice vaccinated with alum/CpG. The addition of CpG to alum decreased OVA IgE responses; however, MPL did not. OVA/alum/CpG increased anti-OVA IgG2a titers in mice housed in paper (1:4.8 × 106) or corncob (1:1.8 × 106) compared with all other OVA-immunized mice. MPL neither enhanced OVA-specific TH1-associated IgG2a levels nor decreased TH2-associated IgE levels under the conditions tested and therefore was removed from subsequent studies. Vaccine adjuvant's influence on peanut-specific antibodies was also measured. Peanut-specific IgE levels were similar among all treatment groups; however, OVA/alum/CpG immunization decreased peanut-specific IgG1 levels compared with those in corncob/OVA/alum mice.
Fig 1Antigen-specific immune responses induced by vaccination and allergen sensitization. Mice in paper or corncob bedding were OVA naive or immunized with OVA/alum, OVA/alum/CpG, or OVA/alum/MPL. A, Antigen-specific serum antibodies were measured by means of ELISA (n = 9-15 mice per group). B, Splenic cytokine responses induced by peanut or OVA were measured by using the multiplex assay (n = 8-10 mice per group). Significant increases are indicated by numbers above bars (P < .05). GMT, Geometric mean titer; PN, peanut.
Splenic T-cell cytokine responses were measured to further assess the influence of alum/CpG on TH1 OVA immunity and evaluate bystander immunity to peanut, regardless of living environment, because bedding type did not globally influence antibody responses (Fig 1, B). Alum enhanced OVA-induced TH2-associated IL-4, IL-5 and IL-13 levels compared with those in naive, peanut-immunized, and OVA/alum/CpG-immunized mice. OVA-specific TH1-associated IFN-γ levels were increased by using alum/CpG compared with naive, peanut alone, and OVA/alum mice. Peanut-induced cytokine responses were not changed by vaccine adjuvants.
Peanut-induced hypothermia and allergy symptoms were measured after completion of OVA vaccination and peanut sensitization. Naive mice exhibited no allergy symptoms or hypothermia. Mice vaccinated with peanut alone or OVA/alum displayed significant hypothermia and clinical symptoms compared with naive mice after challenge (Fig 2, A and B). Mice vaccinated with OVA/alum/CpG exhibited body temperatures not significantly different from those of peanut-naive mice. Exposure to OVA/alum/CpG in paper-housed mice significantly reduced clinical symptoms compared with peanut alone and OVA/alum mice living in corncob bedding. Although there is variability in allergic symptoms within treatment groups, our results suggest that CpG combined with alum indirectly influences host immunity to reduce anaphylaxis because alum/CpG adjuvants were administered on a different day and through a different route than peanut sensitization. In support of our observation, similar intragroup variability has been observed by others after treatment with allergen coadministered with CpG in nanoparticle immunotherapy.
No differences in allergic disease were observed based on living environment. Our results suggest that modification of childhood vaccine adjuvant formulations might provide a method to reduce the development/severity of allergic disease.
Fig 2Adjuvant combination of alum and CpG reduces allergic disease. A and B, C3H/HeOuJ mice were challenged with peanut and monitored for hypothermia (Fig 2, A) and clinical symptoms (Fig 2, B) based on prechallenge phenotype. Statistical differences between groups are indicated by numbers (P < .05). Data were combined from 2 independent experiments (n = 9-15 per group). C and D, OPLS-DA scores plot associated with symptom scores comparing animals that exhibited (Fig 2, C) no change (score 0) and mild disease (score 2) (Fig 2, D) with severe disease (score 5). PN, Peanut.
Multivariate analysis of the binned nuclear magnetic resonance (NMR) metabolomics data differentiated animals based on clinical symptom scores. Animals with a score of 5 were well differentiated from those with scores of 0 and 2 (Fig 2, C and D). The metabolites important to differentiating the groups are listed in Table E4 in this article's Online Repository at www.jacionline.org. Therefore this exploratory analysis suggests that fecal metabolomics might be useful in predicting the severity of the host response to allergens.
In this study we tested the influence of environmental endotoxin and alum adjuvants on development of allergic symptoms. Differences in anaphylactic symptoms were not observed based on bedding or OVA/alum compared with symptoms seen in control mice. Therefore we concluded that neither environmental endotoxin nor the presence of alum, under the conditions used, influenced allergic disease. However, alum/CpG-adjuvanted vaccines administered separately from peanut sensitization reduced allergic symptoms. Exploratory fecal metabolomics were able to identify unique profiles based on living environment and severity of allergic reactions.
According to the “old friends” hypothesis, increased microbial exposure inversely correlates with allergic disease.
National Academies of Sciences, Engineering, and Medicine Finding a path to safety in food allergy: assessment of the global burden, causes, prevention, management, and public policy.
We did not observe differences in allergic disease despite types of animal bedding exhibiting different levels of endotoxin before animal exposure. However, there are still many uncertainties surrounding the influence of environmental endotoxin on allergic disease. Fecal metabolites related to the gut microbiome were identified that differentiated mice based on bedding type. It is possible that the gut microbiota influences host responses to endotoxin and regulates subsequent allergic sensitivity. Additional studies might need to completely separate animals from their feces or use mice receiving antibiotic therapy to reduce gastrointestinal microbial colonization to evaluate the contribution of environmental endotoxin to allergic disease.
Alum's ability to induce antigen-specific TH2 responses and create a bystander response that alters peanut immunity was evaluated. Despite the ability of alum to provide bystander immune activation to newly administered antigens
and our observation that alum enhanced OVA-induced TH2 responses, exposure to alum did not enhance allergy severity in our model. It is possible that our use of cholera toxin (CT) to sensitize mice to peanut might induce potent peanut-specific TH2 responses that cannot be further enhanced by other TH2-skewing adjuvants. Therefore additional work might be required to modify peanut sensitization regimens to allow for proper evaluation of the influence of environmental factors on allergic disease.
We believe that the immune responses modulated by alum/CpG observed in our study are examples of beneficial heterologous effects of vaccination.
Vaccine adjuvants induced unique innate cytokine responses 6 hours after immunization (see Fig E2, A, in this article's Online Repository at www.jacionline.org). Innate serum cytokines induced by gastric peanut sensitization were modified by the vaccine adjuvants administered subcutaneously on the previous day (see Fig E2, B). These data suggest that vaccine adjuvants might alter subsequent immune responses to newly introduced antigens and prospective allergens. Modulation of innate cytokine production might be the basis of vaccines providing heterologous immunity that influence allergic disease. Alum/CpG subsequently induced vaccine-specific TH1 immunity and allergic responses statistically similar to those of naive mice, suggesting that this combination reduced allergic disease severity. However, complete disease protection was not achieved in our study, and CpG might not improve allergy in human infants. Previous reports indicate that CpG does not enhance TH1 cytokines produced by neonatal human cells
Therefore additional research is needed to identify safe and effective adjuvants that induce balanced TH1/TH2 responses in children and might provide indirect protection against allergy.
Broad-spectrum metabolomics using feces collected before peanut challenge provided additional information on disease severity. These metabolites include arginine, formate, glucose, glutamine, methanol, acetate, and methionine, which are consistent with a previous metabolomics study on asthma pathogenesis in human subjects.
In summary, exposure to an alum/CpG-adjuvanted vaccine reduced the severity of peanut hypersensitivity in mice, whereas exposure to endotoxin and alum did not influence allergic symptoms. Fecal metabolomics can provide a method to predict the development, severity, or both of allergic reactions. Additional studies are needed to identify environmental contributions to human allergic disease, their mechanism of action, and prediction of allergy status before an allergic response.
Methods
Mice
Female 4- to 6-week-old C3H/HeOuJ mice (n = 9-15 per group) were purchased from the Jackson Laboratory (Bar Harbor, Me; stock no. 000635) and housed in microisolator cages under specific pathogen-free conditions. Results were obtained from 2 independent experiments and combined. Animal procedures were approved by Duke's Institutional Animal Care and Use Committee.
Experimental design
Mice were separated into 10 groups (Table E1). Five groups were housed in Certified/Irradiated Diamond Soft paper bedding (Harlan Teklad, Indianapolis, Ind), and 5 groups were housed in Enrich-o-cob corncob bedding (Andersons Laboratory, Maumee, Ohio) for 1 month before initiation of vaccinations/allergic sensitization. Paper bedding was replaced 3 times a week to represent a low-endotoxin environment, and the standard endotoxin environment contained corncob bedding that was replaced biweekly. Mice living in paper or corncob bedding included naive mice and mice sensitized to peanut but no other vaccination (peanut alone); mice vaccinated with OVA and alum (alum); mice vaccinated with OVA, alum, and CpG (alum/CpG); or mice vaccinated with OVA, alum, and MPL (alum/MPL) before peanut sensitization. Vaccines were administered subcutaneously beginning 1 day before gastric peanut sensitization. All mice were challenged with peanut and monitored for hypersensitivity reactions on day 36.
Endotoxin measurement
Endotoxin levels were measured in mouse bedding after autoclaving and before animal exposure. Bedding samples were processed, as previously described.
Supernatants were analyzed with the Pierce LAL Chromogenic Endotoxin Quantification Kit (Thermo Scientific, Waltham, Mass). Data are presented as endotoxin units per milligram of bedding.
Vaccinations
Subcutaneous vaccines containing low-endotoxin OVA (Endofit OVA, 10 μg; InvivoGen, San Diego, Calif) and the following adjuvants were administered in 100-μL volumes on days 0, 14, and 28 in sterile USP saline (APP Pharmaceuticals, Schaumburg, Ill): alum (0.5 mg of Al[OH]3; Alhydrogel, InvivoGen), alum/CpG-1826 (100 μg; Vaccigrade, InvivoGen), or alum/MPL (25 μg; Enzo Life Sciences, Farmingdale, NY).
Peanut sensitization
Crude peanut extract was obtained, as previously described.
Peanut sensitization was achieved by means of gastric gavage of peanut (1 mg) and CT (10 μg; catalog no. 100B; List Biologicals, Campbell, Calif) in 200 μL of USP saline on days 1, 8, 15, and 22.
Peanut provocation
Temperature transponders (BioMedic Data Systems, Seaford, Del) were injected subcutaneously before allergen challenge and used to monitor body temperature after peanut challenge. On day 36, mice were challenged intraperitoneally with peanut (250 μg). Body temperature and clinical symptoms were monitored before and every 20 minutes after peanut challenge for 1 hour. Clinical symptoms were reported by using a 0- to 5-point scoring system, as previously described.
The following score system was used to grade hypersensitivity reactions in mice after challenge: 0, no symptoms; 1, repetitive scratching with hind legs near the ear canal or mouth; 2, decreased activity, self-isolation, or swollen snout; 3, motionless for greater than 1 minute or laying prone on stomach; 4, no response to whisker stimulation or decreased or no activity to prodding; and 5, convulsion, death, or temperature decrease of greater than 5°C.
Sample collection
Blood was collected on day 35 by using the submandibular lancet method,
and serum antibody levels were measured by using ELISA. Day 35 fecal pellets were frozen at −80°C for metabolomics.
ELISA
Antigen-specific log2 end point antibody titers were determined by means of ELISA. ELISA plates were coated with 2 μg/mL peanut or OVA and completed as previously described.
Maximal adjuvant activity of nasally delivered IL-1alpha requires adjuvant-responsive CD11c+ cells and does not correlate with adjuvant-induced in vivo cytokine production.
Antigen-specific IgE ELISAs were coated with 5 μg/mL anti-mouse IgE-unlabeled (SouthernBiotech, Birmingham, Ala), and biotinylated antigens were used to determine antigen specificity at 2 μg/mL. The end point titer was calculated as the last sample dilution with a signal 3-fold greater than that of a naive sample at the same dilution.
Innate cytokine analysis
Two groups of mice (n = 10) were immunized subcutaneously with OVA vaccines adjuvanted with alum or alum/CpG on day 0. One group of 8 mice remained OVA naive. Serum was collected from each animal 6 hours after vaccination. Serum was also collected from naive mice at the same time as immunized mice. Five mice from each vaccine group, including naive mice, were sensitized gastrically with peanut/CT 24 hours after OVA vaccination. Serum was collected for a second time from all mice 6 hours after administration of the peanut sensitization dose and analyzed for cytokine responses by using a Luminex multiplex assay (Bio-Rad Laboratories, Hercules, Calif).
Antigen-induced cytokine analysis
Naive, peanut alone, alum, or alum/CpG mice living in the corncob environment because the living environment did not affect humoral immunity were analyzed for peanut- and OVA-induced T-cell cytokine responses after OVA immunization and peanut sensitization. On day 36, spleens were harvested and cultured in vitro with media alone, peanut (200 μg/mL), or OVA (200 μg/mL) for 96 hours by using similarly described methods.
Type B CpG oligodeoxynucleotides induce Th1 responses to peanut antigens: modulation of sensitization and utility in a truncated immunotherapy regimen in mice.
Supernatants were analyzed for cytokine responses by using a Luminex multiplex assay.
Immunology statistical analysis
The unpaired Student t test was used to determine statistical differences between endotoxin levels in bedding samples. ANOVA with Tukey multiple comparisons was used to compare antibody and challenge results. Antigen-induced cytokine responses were calculated by subtracting the picogram per milliliter value from wells that contained cells in media from the picogram per milliliter values of wells stimulated with antigen (peanut or OVA) and then log10-transformed for normalization before analysis with 1-way ANOVA with Tukey multiple comparisons. GraphPad Prism 7 software (GraphPad Software, La Jolla, Calif) was used for all analyses.
Broad-spectrum NMR metabolomics
A subset of animals was chosen from immunologic studies for the exploratory broad-spectrum metabolomics analysis, and sample preparation and data collection and analysis were as described previously.
Briefly, fecal pellets were homogenized in ceramic bead tubes (MagNA Lyser) and D2O (Sigma-Aldrich, St Louis, Mo). Each NMR sample consisted of 50 mg/mL fecal extract with 10% Chenomx ISTD solution (Chenomx, Edmonton, Alberta, Canada) containing 5 mmol/L 4,4-dimethyl-4-silapentane-1-sulfonic acid (chemical shift indicator), 100 mmol/L imidazole (pH indicator), and 0.2% NaN3 (to inhibit bacterial growth). NMR spectra were processed by using the ACD NMR Processor 12.0 (ACD Labs, Toronto, Calif) and binned (0.10-9.00 ppm) into 226 integrated segments (NMR bins), excluding water (4.68-4.88 ppm) and imidazole (7.10-7.50 ppm) by using intelligent bucket integration with a 0.04-ppm bucket width and 50% looseness. Binned data were normalized to the total integral of each spectrum. Before multivariate analysis, normalized binned data were mean centered and Pareto scaled. All of the metabolomics data generated from this study and associated metadata have been uploaded to the public accessible NIH Common Fund Metabolomics Data Repository (http://www.metabolomicsworkbench.org).
Multivariate and statistical analysis of metabolomics data
Descriptive statistics and 2-sided hypothesis tests using continuously binned NMR-binned data were conducted with SAS 9.4 software (SAS Institute, Cary, NC). Hypothesis testing was conducted by using the t test with the Satterthwaite correction for unequal variances. Multivariate data analysis methods (eg, principal component analysis and orthogonal partial least-square discriminant analysis [OPLS-DA]) were used to reduce the dimensionality and enable the visualization of the separation of study groups (SIMCA 14; Umetrics, Umeå, Sweden). These pattern recognition methods are used commonly to analyze high-dimensional multicollinear data, such as metabolomics.
OPLS-DA is a supervised analysis for categorical outcomes and was used to determine the NMR bins that were important for differentiating the phenotypic groups. Loadings plots and variable importance for projection (VIP) plots were inspected along with the VIP statistic, which summarizes the importance of the bin in differentiating phenotypic groups in a supervised multivariate analysis.
Model statistics are reported for OPLS-DA models, including the proportion of variance in the metabolomics data associated with the phenotype (R2X), the proportion of variance in the phenotype explained by the model (R2Y), and the measure of predictive variation of the model assessed by means of 7-fold cross validation (Q2). Bins with either a VIP of 1.0 or greater with a jack-knife CI that did not include 0, a P value of less than .1, or a magnitude of fold change of greater than 2 were determined to be important to differentiating the study groups. Chenomx NMR Suite 8.1 Professional software, which has a concentration library of approximately 350 compounds, was used to match the signals in the bins important to differentiating study groups to metabolites. NMR signals that could not be library matched were classified as unknown. Metabolic pathway analysis was performed by using the MetaCore module in GeneGo software (GeneGo, Philadelphia, Pa).
Fig E1Endotoxin levels are lower in paper compared with corncob bedding. A, Endotoxin levels in bedding before animal exposure and 1 or 3 days after animal exposure. *Significant increase in endotoxin units per milligram compared with paper bedding, as determined by using the unpaired Student t test (P = .0002). B, OPLS-DA score plot showing separation of animals that lived on corncob bedding (blue) from those that lived on paper bedding (green; n = 53, R2X = 0.504, R2Y = 0.597, Q2 = 0.301).
Fig E2Vaccine adjuvant influences innate cytokine response. Mice were vaccinated subcutaneously with alum or alum/CpG on day 0. Half of the immunized mice were sensitized with peanut plus CT (/PN) 24 hours after immunization (day 1). A, Serum cytokine levels measured 6 hours after subcutaneous vaccination. B, Serum cytokine levels measured 6 hours after sensitization. Cytokines were measured by using the Luminex assay. Numbers indicate significant differences between groups by using 1-way ANOVA (P < .05).
C3H/HeOuJ mice were housed in paper or corncob bedding. Mice were naive to peanut, sensitized to peanut only, or peanut sensitized and immunized with OVA/alum, OVA/alum/CpG, or OVA/alum/MPL. A total of 2 independent experiments were performed with 9 to 15 mice per group. Environmental acclimation occurred 4 weeks before immunization. Peanut-induced anaphylaxis was measured 2 weeks after completion of peanut sensitization.
Library-matched metabolites that differentiate animals living in corncob environments (n = 27) from those living in paper environments (n = 26) are shown. P values are based on t tests (assuming unequal variances), and a positive fold change indicates mean peak intensity of corncob bedding is greater than mean peak intensity of paper bedding.
Table E3Potential pathways that differ between living in corncob versus paper environments
No.
Endogenous networks
P value
False discovery rate
1
Glutamic acid pathways and transport
3.010E-12
1.475E-10
2
L-glutamate pathways and transport
8.667E-09
2.123E-07
3
(L)-proline pathways and transport
3.545E-08
5.790E-07
4
Tyrosine pathway
2.895E-05
3.275E-04
5
L-serine pathways and transport
3.342E-05
3.275E-04
6
(L)-alanine pathways and transport
5.543E-05
3.880E-04
7
(L)-phenylalanine pathways and transport
1.141E-04
6.833E-04
8
Methionine pathways and transport
1.255E-04
6.833E-04
9
Amino acid metabolism: asparagine, aspartic acid, arginine metabolism and transport
1.470E-04
7.205E-04
10
L-ornithine pathways and transport
1.727E-04
7.691E-04
The top 10 endogenous networks identified in differentiating the fecal metabolites associated with paper and corncob environments in which the P value is based on distribution of the endogenous metabolite network and the false discovery rate is less than .001 are shown.
Library-matched metabolites that differentiate the severity of an allergen-induced reaction are shown. Metabolite profiles were identified based on comparisons of allergy symptom scores between mice that displayed symptom scores of 0 and 5 and 2 and 5, regardless of environmental risk factors, including living environment and vaccinations.
Supported by the National Institutes of Health (NIH)/National Institutes of Environmental Health Sciences (1R03ES021036-01A1) in collaboration with the NIH Eastern Regional Comprehensive Metabolomics Resource Core (ERCMRC)/National Institute of Diabetes and Digestive and Kidney Diseases (1U24DK097193 to S.J.S.).
Disclosure of potential conflict of interest: B. T. Johnson-Weaver's institution received grant R03 from the National Institute of Environmental Health Sciences (NIEHS) for this work. S. McRitchie's, W. Pathmasiri's, and S. J. Sumner's institutions received a grant from the University of North Carolina at Chapel Hill for this work. C. Chan's institution received a grant from the NIEHS for this work. M. Kulis is employed by the University of North Carolina at Chapel Hill; his institution received grants from the National Institutes of Health (NIH) and Department of Defense (DOD) for other works, the NIH AITC Review Panel and Allertein. A. W. Burks received personal fees from the NIH; is an HAI Study Section review panel member; is a shareholder of Allertein; receives royalties from the American Society for Microbiology; receives payment from Elsevier as a coeditor of an allergy textbook; holds board membership from FARE, the World Health Organization, and Aimmune Therapeutics; receives grants from FARE, the NIH, and the Wallace Research Foundation; receives consultancy fees from Adept Field Solutions, Aimmune Therapeutics, Astellas Pharma Global Development, Biomerica, Evelo Biosciences/Epiva Biosciences, First Manhattan, Genentech, GLG Research, Insys Therapeutics, Intrommune Therapeutics, PPD Development, Regeneron Pharmceuticals, Sanofi US Services, SRA International, Stallergenes, UKKO, and Valeant Pharmaceuticals North America; and received the patents US-09/731375/8153414 Microbial delivery system, US-09/731375/8153414 Microbial delivery system, CA-2241918 HS-103 CIP: Peanut allergens and methods, US- 08/610424 Immunoassay for peanut allergen, EP-96933862.3 HS-103 CAP: Peanut allergens and methods, and US-13-742828/8815251 Microbial delivery system. H. F. Staats' institution received a grant from the NIH for this work, and he was personally paid consultancy fees from Allertein, is employed by Duke University, received grants from the NIH for other works, holds patents issued with his name listed as an inventor but has not received any payments related to the patents, and receives stipend for his work as an editor from ASM. The rest of the authors declare that they have no relevant conflicts of interest.
31437-9&id=gr1.jpg){kind=link}
31437-9&id=gr2.jpg){kind=link}
31437-9&id=fx1.jpg){kind=link}
31437-9&id=fx2.jpg){kind=link}