M cells in follicle-associated epithelium (FAE) are specialized antigen-sampling cells that take up intestinal luminal antigens. Transcription factor Spi-B regulates M-cell maturation, but the molecules that promote transcytosis within M cells are not fully identified. Here we show that mouse allograft inflammatory factor 1 (Aif1) is expressed by M cells and contributes to M-cell transcytosis. FAE in Aif1−/− mice has suppressed uptake of particles and commensal bacteria, compared with wild-type mice. Translocation of Yersinia enterocolitica, but not of Salmonella enterica serovar Typhimurium, leading to the generation of antigen-specific IgA antibodies, is also diminished in Aif1-deficient mice. Although β1 integrin, which acts as a receptor for Y. enterocolitica via invasin protein, is expressed on the apical surface membranes of M cells, its active form is rarely found in Aif1−/− mice. These findings show that Aif1 is important for bacterial and particle transcytosis in M cells.

The Runt-related transcription factor 2 (Runx2) gene encodes the transcription factor Runx2, which is the master regulator of osteoblast development; insufficiency of this protein causes disorders of bone development such as cleidocranial dysplasia. Runx2 has two isoforms, Runx2-II and Runx2-I, and production of each isoform is controlled by a unique promoter: a distal promoter (P1) and a proximal promoter (P2), respectively. Although several studies have focused on differences and similarities between the two Runx2 isoforms, their individual roles in bone formation have not yet been determined conclusively, partly because a Runx2-I-targeted mouse model is not available. In this study, we established a novel Runx2-manipulated mouse model in which the first ATG of Runx2-I was replaced with TGA (a stop codon), and a neomycin-resistant gene (neo) cassette was inserted at the first intron of Runx2-I. Homozygous Runx2-Ineo/neo mice showed severely reduced expression of Runx2-I, whereas Runx2-II expression was largely retained. Runx2-Ineo/neo mice showed neonatal lethality, and in these mice, intramembranous ossification was more severely defective than endochondral ossification, presumably because of the greater involvement of Runx2-I, compared with that of Runx2-II in intramembranous ossification. Interestingly, the depletion of neo rescued the above-described phenotypes, indicating that the isoform-specific N-terminal region of Runx2-I is not functionally essential for bone development. Taken together, our results provide a novel clue leading to a better understanding of the roles of Runx2 isoforms in osteoblast development.

No abstract

No abstract