Statistics from Altmetric.com
We read with great interest the article by Shanahan et al1 describing the roles of environmental conditions, notably co-housing with wild type (WT) littermates, and mouse genetic background in nucleotide-binding oligomerisation domain-containing protein 2 (NOD2)-dependent production of anti-microbial peptides in the mouse intestine. These authors demonstrate that expression, translation and anti-microbial activity of α-defensins are independent of NOD2.1 Robertson et al2 recently confirmed that housing conditions rather than NOD2 status influenced intestinal microbiota composition. Shanahan et al address the question whether an increase in the number of Paneth cells could compensate for a NOD2-dependent reduction in the level of defensin production in Paneth cells. Using a combination of hematoxylin and eosin staining (to assess crypt numbers), immunohistochemistry (using anti-lyzozyme staining) and flow cytometry (sorting for expression of lyzozyme and lack of expression of CD45—a haematopoietic cell marker) of the entire ilea of WT and NOD2-deficient mice showed NOD2 status did not influence Paneth cell numbers.1
CD24, a heavily glycosylated protein marker of intestinal crypt stem cells and Paneth cells, is upregulated in inflammatory bowel disease (IBD).3 ,4 Sato et al3 demonstrated that the combination of CD24hi and increased side scatter on flow cytometry can be used to sort Paneth cells from mouse intestinal stem cell cultures (‘organoids’).
Recently, a thymine guanine (TG) deletion polymorphism (rs3838646) in the 3′ untranslated region, which reduces the stability of CD24 mRNA, has been associated with susceptibility to Crohn's disease (CD) (OR 1.61) in addition to protection against chronic Hepatitis B Virus infection, systemic lupus erythematosus and multiple sclerosis.5 ,6
Our aims were to assess the influence of NOD2 status on Paneth cell numbers (by CD24-based flow cytometry) using mouse intestinal organoids and to determine the role of the CD24 gene polymorphism (rs3838646) in the genetic susceptibility to IBD in the high-incidence Scottish population characterised by low allelic frequency of the ‘common’ CD-associated NOD2 variants.7
Small intestinal stem cell crypts were isolated from C57BL/6 NOD2WT and NOD2KO mice housed separately in specific pathogen-free conditions.3 Intestinal crypts of NOD2WT and NOD2KO mice (expressing green fluorescent protein controlled by the NOD2 promoter) were maintained in Matrigel (Becton, Dickison & Company) and organoid medium (Advanced DMEM/F12, Glutamax 100×, Hepes 1 M 100×, Pen/Strep 100×, B27 Supplement 50×, N2 Supplement 100× (all from Invitrogen) and N-acetylcysteine 500 mM 400× (Sigma-Aldrich)) containing R-Spondin-conditioned medium (prepared from HEK293T cells expressing R-Spondin, provided by Calvin Kuo, Stanford), murine recombinant Noggin (Peprotech) and murine recombinant epidermal growth factor (Invitrogen) as described by Clevers et al.3 Medium was changed three times weekly and organoids were passaged once weekly. For flow cytometry, organoids were isolated from Matrigel using cold organoid medium (P1000 pipette) followed by trypsinisation to disrupt the intestinal crypts before resuspension in fluorescence-activated cell sorting buffer. There was no difference in Paneth cell numbers (CD24hi, side scatter +) between NOD2WT and NOD2KO intestinal organoids (3.83% vs 3.90%, figure 1).
Next, we performed a genetic association analysis of the rs3838646 polymorphism in 1038 adult IBD patients, 351 childhood-onset IBD (<17 years at diagnosis) and 940 healthy controls (as described previously).8 Due to the presence of a CD24 pseudogene on the Y chromosome, only female IBD patients (n=737) versus female controls (n=498) were analysed to avoid non-specific amplification of the TaqMan assay. Our study had 80% power to detect association with OR 1.46 and OR 1.62 (with IBD and CD, respectively, using Quanto software (http://hydra.usc.edu/gxe)). Genotype frequencies and allelic frequencies did not differ between IBD/CD patients and controls (table 1).
In conclusion, using intestinal stem cell cultures from NOD2WT and NOD2KO mice, we have demonstrated by means of CD24-based flow cytometry that NOD2 status does not influence Paneth cell numbers. In the high-incidence Scottish IBD population, the CD24 gene polymorphism (rs3838646) is not an important determinant of genetic susceptibility to IBD/CD.
Correction notice This article has been corrected since it was published Online First. The Funding statement has been amended.
Acknowledgements The authors would like to thank Dr S Middendorp, Prof E Nieuwenhuis and Prof H Clevers (Utrecht, the Netherlands) for their help and advice in setting up the intestinal organoid cultures.
Contributors JVL: study design, performed experiments, wrote manuscript. KG: study design, performed experiments, manuscript preparation. PH: genotyping, data analysis, manuscript preparation. RKR: patient and data collection, manuscript preparation. HED: patient and data collection, genotyping quality control, data analysis. JS: patient and data collection, study design, manuscript review. AMG: supervisor of experiments, study design, manuscript preparation. DJP: study design, supervisor of experiments, manuscript preparation. DCW: study design, genotyping, patient and data collection, manuscript preparation.
Funding JVL was supported by a Canadian Institutes for Health Research (CIHR)/Canadian Association of Gastroenterology Fellowship (234622), the University of Toronto Edward Christie Stevens Fellowship, Nellie L Farthing Fellowship, the William S Fenwick Fellowship, the Chisholm Memorial Fellowship, the Miriam Neveren Memorial Award and the Joseph M West Family Memorial Fund, and was the holder of the 2011 CIHR Bisby Fellowship. DP is a Howard Hughes International Scholar and is supported by the CIHR and Canadian Crohn's and Colitis Foundation (CCFC). PH was supported by a Medical Research Council project grant for PICTS to DW (G0800675), by Schering-Plough and the GI/Nutrition Research Fund, Child Life and Health, University of Edinburgh. RKR is supported by a NHS Research Scotland career fellowship award.
Competing interests None.
Ethics approval COREC.
Provenance and peer review Not commissioned; internally peer reviewed.
Data sharing statement CD24 genotyping: only female patients and controls were analysed due to pseudogene on the Y chromosome.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.