The reason for the efficient Cldn11 induction in BMDM is unclear, although M-CSF, used to generate BMDM, and IL-4 have been shown before to co-regulate certain genes [30]. A summarized gene expression pattern of all adherence and tight junction proteins in macrophages is provided (See summary in Table 2, right columns). Although IL-4 significantly increases the mRNA levels of claudin-1, 2 and 11, this does not result in a detectable

expression of these proteins in macrophages. As a matter of fact, no reports of claudin protein expression in HTS assay macrophages exist up to now, in contrast to related cell types such as LCs and DCs. Possibly, the claudin protein expression levels in macrophages are under the detection limit of the antibodies currently used. Alternatively, we cannot exclude that post-transcriptional, such as poor

mRNA stability, and/or post-translational regulatory mechanisms preclude high claudin levels in macrophages. For example, during epithelial reorganization, claudins are ubiquitylated and undergo degradation in the lysosomes [31]. A similar mechanism might be at play in macrophages, especially if the claudins are not engaged in TJ formation. In this respect, one could imagine that claudin proteins are stabilized in vivo when intimate interactions between macrophages and epithelial cells are formed. This could help to bring macrophages in close contact with epithelial cells or with other macrophages, a phenomenon that could be relevant in several situations: (1) in tumours where Sirolimus mw fusion between macrophages and carcinoma cells might occur [32], (2) during wound healing where macrophages have to integrate in the epithelial sheet of the skin [33] and (3) during granuloma formation and the foreign body reaction where close contacts between macrophages have to be initiated to promote their fusion [29]. Interestingly, Lenzi et al. [34] reported the expression of cadherins and the tight junction–associated protein occludin during the PTK6 process of granuloma closure. Yet, the lack of claudin proteins in our assays with IL-4-treated macrophages does not preclude their use as marker genes. Indeed, the macrophage activation status in a given pathological

condition is often evaluated by the detection of M1 versus M2 signature genes [4, 25, 26, 35]. Testing different M2 activators identified TGF-β as the most potent inducer of Cldn1 gene expression in macrophages. This finding is reminiscent of TGF-β’s central role in upregulating claudin-1 expression during IL-4-/GM-CSF-treated bone marrow cultures, ultimately giving rise to Langerhans cells [18]. The association of claudin-1 mRNA with the M2 activation status was further confirmed in vivo where high levels of Cldn1 induction were observed in TAM subsets from two mammary carcinoma models and in splenic macrophages isolated from the chronic infection stage of T. congolense infections. In both models, the implication of TGF-β seems plausible.