The A/B domain of truncated retinoic acid receptors can block differentiation and promote features of malignancy.

Recently, we discovered that stable introduction of a carboxyl-terminally truncated retinoic acid receptor gamma (tRAR gamma) into an epidermal keratinocyte line blocked the ability of these cells to differentiate, as judged by their failure to express late markers of squamous differentiation. We now demonstrate a correlation between the level of residual endogenous RAR activity of tRAR gamma-expressing keratinocyte lines and degree of terminal differentiation. Mutagenesis studies localize the effects to the A/B subdomain of the truncated receptor. Despite tRAR gamma's capacity to interfere with RAR-mediated transactivation of retinoic acid response elements (RAREs) in keratinocytes, the effects of the truncated receptor are independent of its ability to bind DNA and directly interact with endogenous RARs. tRAR alpha also inhibits RARE-mediated gene expression in keratinocytes, even though its full-length counterpart enhances RARE activity in these cells. Intriguingly, both tRAR gamma and RAR gamma suppress keratin promoter activity in epidermal cells, although for tRAR gamma, the effect is mediated through the A/B domain whereas for RAR gamma, the effects require DNA binding. Taken together, these findings suggest that the truncation allows for new and aberrant interactions with transcriptional proteins/cofactors that participate in governing RARE activity. This discovery may have relevance in tumorigenesis, where genetic lesions can result in mutant RARs or in loss of receptor expression.

Terminal differentiation in keratinocytes involves positive as well as negative regulation by retinoic acid receptors and retinoid X receptors at retinoid response elements.

Terminal differentiation of epidermal keratinocytes is inhibited by 1 microM retinoic acid, a concentration which induces differentiation in a number of cell types, including F9 teratocarcinoma cells. The molecular basis for these opposing retinoid responses is unknown, although retinoic acid receptors (RARs) and retinoid X receptors (RXRs) have been detected in both cell types. When F9 cells are stably transfected with a truncated RAR alpha lacking the E/F domain necessary for ligand binding and RAR/RXR dimerization, action at retinoid response elements is suppressed and cells produce a retinoic acid-resistant phenotype; i.e., they are blocked in differentiation (A. S. Espeseth, S. P. Murphy, and E. Linney, Genes Dev. 3:1647-1656, 1989). If retinoid receptors influence epidermal differentiation only in a negative fashion, then suppression of transactivation at retinoid response elements would be expected to enhance, rather than block, keratinocyte differentiation. In this study, we show that surprisingly, even though constitutive expression of an analogous truncated RAR gamma in keratinocytes specifically suppressed transactivation at retinoid response elements, keratinocytes were blocked, rather than enhanced, in their ability to undergo morphological and biochemical features of differentiation. These findings demonstrate a direct and hitherto unrecognized role for RARs and RXRs in positively as well as negatively regulating epidermal differentiation. Additionally, our studies extend those of Espeseth et al. (Genes Dev. 3:1647-1656, 1989), indicating a novel RAR function independent of the E/F domain.

Phosphorylation of integrin in differentiating ts-Rous sarcoma virus-infected myogenic cells.

The differentiation of primary myogenic cultures requires the attachment of the cells to an extracellular matrix substrate using an integrin family receptor. These integrin receptors can be phosphorylated on both their alpha and beta chains, and it has been postulated that phosphorylation regulates the receptor function. Quail myogenic clones transformed with ts-LA24A differentiated into mature myotubes following a temperature shift to nonpermissive temperature which inactivates the viral src kinas. Phosphorylation of integrin beta-1 chain and of at least one alpha chain was detected on both serine and tyrosine. An additional alpha chain(s) with a mobility similar to alpha 5 was not phosphorylated at either temperature. Following the induction of differentiation by a temperature shift, there was a marked decrease in integrin phosphorylation of both alpha and beta integrin chains. This decrease was more prominent for serine than for tyrosine, suggesting that src could not be the only kinase involved. The drop in integrin phosphorylation correlated with the initiation of differentiation, suggesting that integrin phosphorylation could be at least part of the mechanism by which myogenic differentiation is blocked by v-src.

Cellular partitioning of beta-1 integrins and their phosphorylated forms is altered after transformation by Rous sarcoma virus or treatment with cytochalasin D.

A sequential extraction procedure of 3-[(3-cholamidopropyl)-dimethylammonio]-1-propane sulfonate (CHAPS) buffer followed by RIPA or Laemmli sample buffer was developed to define two distinct subpopulations of beta-1 integrins in primary chicken embryo fibroblasts. Extraction of cells in culture revealed an association of adhesion plaque-localized integrin with the CHAPS-insoluble fraction. Phosphorylated integrins were found in both fractions, but the specific phosphorylation was 12-fold higher in the CHAPS insoluble fraction. The phosphorylation was evenly distributed between phosphoserine and phosphotyrosine. Transformation by Rous sarcoma virus caused a redistribution of integrin to rosettes and an increase in total integrin phosphorylation. Treatment with cytochalasin D caused a redistribution of the adhesion plaque-associated integrin into lacelike structures and reduced the level of integrin phosphorylation. These treatments also caused an altered distribution of phosphorylated integrin between the CHAPS soluble and insoluble fractions. These results suggest a role for integrin phosphorylation in the assembly and disassembly of cellular adhesion structures.

Analysis of adenovirus early and late gene expression in cultured epidermal keratinocytes.

Previous studies have shown that expression of adenovirus type 2 (Ad2) is restricted in epidermal keratinocytes in what appears to be a differentiation specific manner. We have analyzed the relationship between keratinocyte differentiation and Ad2 early and late gene expression. Cultured epidermal keratinocytes infected with Ad2 were fractionated in density gradients of Ficoll 400 to enrich for populations of nondifferentiated cells and differentiated cells. Analysis of these populations revealed that both populations supported early Ad2 gene expression but restricted Ad2 late gene expression. The restriction to late gene expression differed in the two cell populations: Nondifferentiated keratinocytes did not support production of high levels of Ad2 capsid proteins, whereas differentiated keratinocytes supported synthesis of Ad2 capsid proteins but restricted Ad2 expression at a later step that normally leads to production of high titers of progeny virus. The changing restriction to Ad2 expression during keratinocyte differentiation may have resulted from changes in cellular components that play a role in cell differentiation.

Evidence for two points of restriction in the expression of adenovirus type 2 in cultured epidermal keratinocytes.

Cultures of epidermal keratinocytes contain two populations of cells, a basal undifferentiated population and a suprabasal terminally differentiated population. When exposed to wild-type adenovirus type 2 (wtAd2), the suprabasal cells are positive by immunofluorescence for capsid antigen and exhibit cytopathic effects (CPE) (R.F. LaPorta, and L.B. Taichman, Virology 110:137-146, 1981). The basal cells, although infected, are not positive for capsid antigen and do not display CPE. Despite CPE and capsid antigens in suprabasal cells, yields of virus from the entire culture are very low (10 PFU per cell). These observations suggest that Ad2 expression is restricted at different times in the viral life cycle in basal and suprabasal cells. To test this hypothesis, we isolated host range (hr) mutants of Ad2 on two lines of squamous cell carcinoma (SCC) keratinocytes which were shown to be restrictive for wtAd2 replication. The hrAd2 mutants produced high yields of progeny virus in epidermal cell cultures (500 to 600 PFU per cell). However, the pattern of CPE induction in these cultures was like that produced by wtAd2, i.e., basal cells were CPE negative and suprabasal cells were CPE positive. The high yield of hrAd2 progeny indicated that the restriction present in suprabasal cells was overcome. However, the failure of hrAd2 mutants to induce CPE in basal cells indicated that the hrAd2 mutants remain restricted in the basal population and supported our hypothesis that a second and distinct restriction exists in basal keratinocytes.

Epithelium-specific response of cultured keratinocytes to infection with adenovirus type 2.

Adenoviruses are pathogenic for certain stratified squamous epithelia. The sites most frequently involved are the upper respiratory tract and oropharynx. Adenovirus infections of the epidermis are quite rare. We examined the virus-cell interactions of adenovirus type 2 (Ad2) and cultured human keratinocytes grown from a variety of body sites. Our intent was to explore the nature of the apparent epithelium-specific susceptibility to Ad2. In brief, we found that in vitro viral susceptibility of the keratinocytes could be reliably predicted based on whether the cells originated from an epidermal or oropharyngeal surface. Ad2 proceeded through a complete vegetative cycle when used to infect cultured keratinocytes from oropharyngeal sites (e.g., gingiva and soft palate). In contrast, Ad2 infection was severely restricted in keratinocytes from epidermal sites (e.g., foreskin, abdomen, and buttock). These results demonstrate that the in vitro response to infection with Ad2 reflects in vivo tissue-specific susceptibility. In vivo, cervical epithelium is rarely infected with Ad2 and yet in culture, cervical keratinocytes were fully permissive for Ad2 replication. We propose that the permissive or nonpermissive response to Ad2 may be regulated by a particular aspect of cell phenotype. Because the permissive responses seen in this study were all generated in keratinocytes from mucosal sites, it is possible the in vitro response to Ad2 reflects inherent differences between mucosal and epidermal keratinocytes.