Effect of estrogen administration on rat liver 2-5A synthetase activity.

Interferon-induced 2-5A synthetase is also present in various cells and tissues in the absence of any interferon treatment. The activity of this enzyme, which synthesizes a series of oligoadenylates, ppp(A2'p)n5'A (collectively referred to as 2-5A), was previously shown to vary with the growth status of liver tissue i.e., it decreased before and during the peak of DNA synthesis activity induced in rat liver by a two third hepatectomy. In the course of studies aimed at testing the hypothesis that 2-5A synthetase activity might exert negative control on normal cell growth and multiplication, we show here that a treatment of ovariectomized rats with a single dose of estradiol-17beta (100 micrograms/100 g body weight) induced a transient increase in the [3H]thymidine labelling index in the liver after 24 h and markedly decreased the 2-5A synthetase activity. A time course study revealed that 2-5A synthetase activity started to decrease after 3 h, reaching a minimal value (10% of the control level) after 12 h, then slowly increased to come back to control level at 48 h. These results, together with our similar data on regenerating liver, suggest that low 2-5A synthetase activity is permissive for acquisition of proliferative 'competence' by G0 cells.

Specific internalization of basal membrane domains containing the integrin alpha 6 beta 4 in dispase-detached cultured human keratinocytes.

The integrin alpha 6 beta 4 is polarized towards the basal side of basal keratinocytes and helps anchor these cells to the basement membrane components. We have found that cultured human epidermal keratinocytes, when detached from their culture substratum, as for grafting, using the enzyme dispase, rapidly internalize the basal membrane domains containing the integrin alpha 6 beta 4, while integrins of the very late antigen subtype remain on the cell surface. Detachment and incubation at 4 degrees C prevent this internalization, as well as the contraction of the detached sheet area. Subsequent incubation at 37 degrees C initializes this contraction and allows the basal integrin alpha 6 beta 4 to be internalized. We took advantage of this blockage to label upon detachment using immunogold techniques, the alpha 6 subunit present on the basal cell surface; then we studied its internalization with the electron microscope. This internalization pathway differs from classical receptor-mediated endocytosis, and intermediate filaments might possibly play a role in this process. Interestingly, 1 h after their internalization from the basal membrane, a third of the gold particles labeling the alpha 6 subunit was found between lateral membranes of basal cells, strongly suggesting that the integrin alpha 6 beta 4 can be partly recycled to the cell surface in these conditions.

The tetraspanin CD9 associates with the integrin alpha6beta4 in cultured human epidermal keratinocytes and is involved in cell motility.

Integrins are involved in several ways in keratinocyte physiology, including cell motility. CD9 is a member of the tetraspanin protein family which is found in association with other transmembrane proteins like the integrins. CD9 is expressed in the epidermal tissue, but this expression is not regulated by differentiation. The present work focuses on association of CD9 with the integrin alpha6beta4 in keratinocytes. In vivo, CD9 does not co-localize with alpha6beta4, and is not internalized with the integrin upon basal detachment with dispase. In vitro, CD9 is found partly in co-localization with alpha6beta4 and is internalized with the integrin after keratinocyte detachment with dispase. Using blocking antibodies in a phagokinetic tracks assay, we show that CD9, and to a lesser extent alpha6beta4, but not the tetraspanin CD82, promote motility of subconfluent keratinocytes on collagen I. Our observations also suggest that CD9 is involved in the formation of lamellipodia. We also report that the phorbol ester TPA has no effect on CD9 expression and association with alpha6beta4, but increases keratinocyte motility, possibly through modulation of integrin subunits expression, or through upregulation of collagenase-1 expression. Together, these results confirm that CD9 associates with alpha6beta4 in cultured keratinocytes, possibly in order to regulate the function of the integrin, and that CD9 is involved in keratinocyte motility on collagen. The data suggest that regulation of adhesion characteristics by CD9 in keratinocytes may play a role in epidermal repair.

Basal detachment of the epidermis using dispase: tissue spatial organization and fate of integrin alpha 6 beta 4 and hemidesmosomes.

Dispase has been utilized to produce basal detachment of the epidermis of human skin biopsies and to study the consequences induced afterwards during incubations of the detached tissue. Spatial reorganization of the epidermis is observed under these conditions and is characterized by disappearance of the typical basal keratinocyte layer. Immunofluorescent labelings reveal upward migration of several cells exhibiting the basal phenotype between suprabasal differentiating keratinocytes and demonstrate progressive intracellular expression of hemidesmosomal components: the integrin alpha 6 beta 4 and two plaque components, the 230-kDa bullous pemphigoid antigen and HD1, a 500-kDa protein. Using electron microscopy and immunogold techniques, we demonstrate that the hemidesmosome-containing basal membrane domains enter the cell cytoplasm after detachment of the epidermal tissue. Partial recycling of internalized hemidesmosomal components is also suggested. Our findings illustrate the processing of released hemidesmosomes in detached basal keratinocytes and suggest some heterogeneity between basal cells migrating towards a suprabasal position and those remaining in the basal layer. These results suggest that the dispase-detached epidermis is a self-remodeling tissue in which basal keratinocytes' and tissue's polarities observed in the anchored epidermis are progressively changing.

A simple reconstructed human epidermis: preparation of the culture model and utilization in in vitro studies.

The preparation of a reconstructed human epidermis is described with examples of its utilization in in vitro studies. The model was obtained by culturing normal human keratinocytes at high cell density for 14 days in serum-free and high calcium (1.5 m M) medium on an inert polycarbonate filter at the air-liquid interface. These stratified cultures showed histological features similar to those observed in vivo in the epidermis: a proliferating basal layer and differentiating spinous, granular, and cornified layers. Electron microscopy illustrated lamellar bodies, junctions and keratohyalin granules. Immunofluorescent localization of epidermal markers (keratins 14 and 10, involucrin and filaggrin) revealed typical differentiation. This in vitro reconstructed tissue was used in studies of toxic effects of chemicals. The modelled tissue showed progressive cytotoxicity of a skin irritant (benzalkonium chloride) and a sensitizer (dinitrochlorobenzene) as assessed by MTT assay. Moreover, differential release of interleukin-1alpha and interleukin-8 were measured after 20 h of incubation allowing the irritant to be distinguished from the sensitizer. Permeation studies indicated efficient barrier function of the reconstructed epidermis, as well as metabolizing properties towards hormones. This model can be custom-made and is potentially useful for studies involving keratinocytes in the epidermis, in basic science, dermatology or toxicology.

In Vitro models of epidermal differentiation.

The differentiation program followed by the epidermal keratinocyte in skin is intended to continuously produce and maintain a cornified layer made of fully keratinized cells. This outer layer of the body provides a certain protection against external pathogens and chemical or physical agents, together with a barrier that prevents loss of body fluids. Considerable knowledge of epidermal differentiation and understanding of its regulation has progressively emerged from the availability of keratinocyte cultures, and from the consecutive possibility of unlimited in vitro experimentation. This short review briefly presents the main current in vitro models of epidermal differentiation and emphasises their advantages of pitfalls when studying particular steps of the differentiation program or analyzing their regulation.

NRAGE, a p75NTR adaptor protein, is required for developmental apoptosis in vivo.

NRAGE (also known as Maged1, Dlxin) is a member of the MAGE gene family that may play a role in the neuronal apoptosis that is regulated by the p75 neurotrophin receptor (p75NTR). To test this hypothesis in vivo, we generated NRAGE knockout mice and found that NRAGE deletion caused a defect in developmental apoptosis of sympathetic neurons of the superior cervical ganglia, similar to that observed in p75NTR knockout mice. Primary sympathetic neurons derived from NRAGE knockout mice were resistant to apoptosis induced by brain-derived neurotrophic factor (BDNF), a pro-apoptotic p75NTR ligand, and NRAGE-deficient sympathetic neurons show attenuated BDNF-dependent JNK activation. Hair follicle catagen is an apoptosis-like process that is dependent on p75NTR signaling; we show that NRAGE and p75NTR show regulated co-expression in the hair follicle and that identical defects in hair follicle catagen are present in NRAGE and p75NTR knockout mice. Interestingly, NRAGE knockout mice have severe defects in motoneuron apoptosis that are not observed in p75NTR knockout animals, raising the possibility that NRAGE may facilitate apoptosis induced by receptors other than p75NTR. Together, these studies demonstrate that NRAGE plays an important role in apoptotic-signaling in vivo.

Basal cell adhesion to a culture substratum controls the polarized spatial organization of human epidermal keratinocytes into proliferating basal and terminally differentiating suprabasal populations.

The contribution of adhesion to an extracellular matrix in the polarized spatial organization of keratinocytes was studied in dispase-detached cultures stored as floating sheets. Proliferating and terminally differentiating cell populations were, therefore, localized on tissue sections by their DNA-synthesizing ability and involucrin immunostaining, respectively. A progressive reorganization was induced from superposed proliferating and differentiating layers into clusters exhibiting differentiating cells on the outside. Measurements of proliferation and terminal differentiation in detached cultures revealed the progressive disappearance of proliferating cells, followed by an increase in involucrin-positive cells. Attempts to block the spatial reorganization by the addition of components of the extracellular matrix remained unsuccessful. These results suggest that basal anchorage is responsible for the superposition of proliferating and differentiating cells in keratinocyte cultures. They afford new arguments for the induction of terminal differentiation in non-adhesive keratinocytes which exhibit a concomitant modification of cell shape.

(2'-5')Oligoadenylate in rat liver: modulation after partial hepatectomy.

(2'-5')Oligoadenylate synthetase [(2'-5')A synthetase], which synthesizes a series of oligoadenylates ppp-(A2'p)n5'A [collectively referred to as (2'-5')A], has been described previously in rat liver cells, where its concentration varied with the growth status of this organ--i.e., it decreased during the early phase of rat liver regeneration after partial hepatectomy. Because double-stranded RNA, the only known activator of this enzyme, has been detected in rat liver nuclei, (2'-5')A synthesis could occur in this tissue in vivo. Analysis of rat liver cell extract after HPLC by the endonuclease-based radiobinding assay revealed several components with retention times similar to (2'-5')A trimer- and tetramer-like material. A further characterization of these compounds by their susceptibility to alkaline phosphatase and snake venom phosphodiesterase, their resistance to micrococcal nuclease, and their ability to activate an endonuclease indicated the natural occurrence of oligonucleotides indistinguishable from authentic (2'-5')A in rat liver cells. Using the combination of the radiobinding assay and a simplified (2'-5')A extraction procedure that does not involve HPLC, we further show that the early drop of (2'-5')A synthetase activity during rat liver regeneration was accompanied by a similar decrease in intracellular (2'-5')A concentration. The three characteristic phases of the (2'-5')A synthetase kinetics during the first 40 hr of liver regeneration were mimicked by the kinetics of the synthesis of the (2'-5')A oligonucleotides themselves: between 6 and 20 hr after hepatectomy, there was a sharp decrease in (2'-5')A concentration; between 20 and 24 hr, the concentration of (2'-5')A reached a minimum; at 36 hr or after the first wave of DNA synthesis (the major event of liver regeneration), the (2'-5')A concentration returned to normal. In this characterization of the (2'-5')A oligonucleotide family in a functional tissue of an animal that had not been previously treated with interferon or infected with virus, the data are compatible with a physiological role of the (2'-5')A system acting as an intracellular component of the regulatory mechanisms leading to cell proliferation or differentiation.