As initial step to determine the localization of the glucocorticoid receptor (GR) in primary human keratinocytes we used immunohistochemistry. Primary human keratinocytes were treated with 1 μM of dexamethasone (Dex) for 30 minutes and upon fixation stained with GR-specific antibody to determine the cellular localization of the receptor. In addition, we used α-catenin specific antibody to mark the plasma membrane. As expected, GR was found in the cytoplasm and nucleus of the keratinocytes (). Interestingly, we also found a fraction of GR that was localized on the plasma membrane (; C). Upon further examinations we observed that GR co-localized, in part, with α-catenin (; C). This suggests that GR may be associated with adherens junctions (AJ).
Co-localization of the GR and α-catenin to the plasma membrane of keratinocytes.
Next, we performed cellular fractionations and Western blotting to establish the presence of GR in various cellular compartments. To test how presence of hormone influences GR localization primary human keratinocytes were incubated in the presence or absence of Dex. The purity of the fractions were confirmed as follows: histone H3 was found only in nuclear fractions whereas it was absent from cytoplasmic and membranous (). Conversely, IkBa was found only in cytoplasmic fractions whereas it was absent from either nuclear or membranous. Interestingly we found GR present in all three cellular compartments (). As expected, nuclear presence of GR markedly increased upon hormone treatment, whereas cytoplasmic significantly decreased, confirming hormone-dependent translocation from cytoplasm to the nucleus. Furthermore, GR maintained its presence in membranous fraction in both conditions. Membranous GR was decreased in the presence of hormone, suggesting that membranous GR may also contribute to nuclear GR. Importantly, continuous presence of GR in the membranous fraction supports the notion of the association of GR to the plasma membrane.
GR is bound to α-catenin in the plasma membrane of keratinocytes.
GR is known to inhibit wound healing by disrupting migration of keratinocytes 
. The co-localization of GR and α-catenin raised an intriguing possibility that membranous GR may be associated with adherens junctions (AJ), especially because cell-cell interactions play an important role in maintaining both migration and proper skin barrier formation. To test GR interaction with AJ components, we performed co-immunoprecipitation studies. Interestingly, GR was found in a complex bound to α-catenin (). Next, we performed the converse experiment. This time we used GR-specific antibody to pull down GR-bound proteins from the membranous fraction. As expected, GR was found present in both fractions, although it showed slight decrease upon incubation with hormone (). Next we probed the same membrane with α-catenin specific antibody and found presence of α-catenin in both fractions. GR-associated α-catenin shows slight increase in the fraction obtained from cells incubated with hormone. Further, we examined whether GR interacts with α-catenin in human epidermis. Upon incubation of human skin in the presence or absence of Dex, the epidermis was separated from the dermis, proteins were isolated, followed by immunoprecipitation with α-catenin specific antibody to pull down α-catenin bound proteins. GR was found in a complex bound to α-catenin recapitulating the findings observed in primary human keratinocytes (). We conclude that the interaction between GR and α-catenin occurred in the presence and absence of the hormone, suggesting GR association to adherens junctions in mediating constitutive, non-genomic effects.
Our initial observation leads to conclusion that GR localizes to the plasma membrane of keratinocytes as an element of adherens junctions, in association to α – catenin.