Keratinocytes as depository of ammonium-inducible glutamine synthetase: age- and anatomy-dependent distribution in human and rat skin

Abstract

In inner organs, glutamine contributes to proliferation, detoxification and establishment of a mechanical barrier, i.e., functions essential for skin, as well. However, the age-dependent and regional peculiarities of distribution of glutamine synthetase (GS), an enzyme responsible for generation of glutamine, and factors regulating its enzymatic activity in mammalian skin remain undisclosed. To explore this, GS localization was investigated using immunohistochemistry and double-labeling of young and adult human and rat skin sections as well as skin cells in culture. In human and rat skin GS was almost completely co-localized with astrocyte-specific proteins (e.g. GFAP). While GS staining was pronounced in all layers of the epidermis of young human skin, staining was reduced and more differentiated among different layers with age. In stratum basale and in stratum spinosum GS was co-localized with the adherens junction component beta-catenin. Inhibition of, glycogen synthase kinase 3beta in cultured keratinocytes and HaCaT cells, however, did not support a direct role of beta-catenin in regulation of GS. Enzymatic and reverse transcriptase polymerase chain reaction studies revealed an unusual mode of regulation of this enzyme in keratinocytes, i.e., GS activity, but not expression, was enhanced about 8-10 fold when the cells were exposed to ammonium ions. Prominent posttranscriptional up-regulation of GS activity in keratinocytes by ammonium ions in conjunction with widespread distribution of GS immunoreactivity throughout the epidermis allows considering the skin as a large reservoir of latent GS. Such a depository of glutamine-generating enzyme seems essential for continuous renewal of epidermal permeability barrier and during pathological processes accompanied by hyperammonemia.

Figure 1. Presence of GS in human foreskin of children and regional variations in its distribution revealed in co-localization studies with GFAP, or with SMAA.

(A–D) Expression of GS (A) and GFAP (B) and merged micrograph of GS and GFAP (C) immune reaction in human foreskin sections. (D–F) Expression of GS (D) and SMAA (E) and merged micrograph of GS and SMAA immune reaction (F) in a human skin section. The line connecting the D–F clearly shows strong expression of SMAA (E) in stratum granulosum and its absence of stratum corneum. The latter is strongly stained for GS (D). GS and SMAA merge at the interface of stratum corneum boarder and stratum granulosum (F). Binding of specific antibodies was visualized with FITC-conjugated goat anti-rabbit IgG and Cy3-conjugated goat anti-mouse IgG. Blue arrows point to stratum corneum, magenta arrows to stratum basale white arrows to fibroblasts and yellow arrows to small and middle-sized vessels. Scale bar: A–D, H 200 μm; E–G 100 μm.

Figure 2. Distribution of GS and β-catenin in normal senile human temple skin.

(A) hematoxiline staining of a skin sample. Stratum basale can be identified by cells with prominent nuclei. (B) GS is localized throughout the viable part of the epidermis. The highest staining intensity can be detected at the interface of stratum granulosum and stratum corneum. (C) Localization of β-catenin in lower parts of the epidermis. β-catenin shows a strong membraneous and cytoplasmic staining in stratum basale and in stratum spinosum. Arrows in (C) point to the presence of catenin in the nucleus of keratinocytes. Scale bar 25 μm.

Figure 3. Expression of GS in newborn rat scalp sections CLS microscopy of skin preparations stained for GS.

(A) Representative of serial optical sections obtained by CLS microscopy of newborn rat skin stained for GS (green). (B) Phase contrast micrograph of the field shown in (A). (C) Wide-field fluorescence microscopy of fibroblasts in human skin. (D, H) Representative of serial optical sections obtained by CLS microscopy of cultured human fibroblasts stained for GS. GS was detected using rabbit antiserum and FITC-conjugated goat anti-rabbit IgG. (E, F) Representatives of serial optical sections obtained by CLS microscopy of hair follicle in newborn rat skin section stained for GS. (G) Phase contrast micrograph of the field shown in E and F. Blue arrows point to epidermis, magenta arrows to the keratinocytes forming the stratum basale of the epidermis, white arrows to fibroblasts, green arrows to the outer sheet of the hair follicle, red arrows to the hair channel. Scale bar in (A, B, D, H) 20 μm; in (C) 100 μm; in (E, F, G) 40 μm.

Figure 4. Age-dependent distribution of GS and GFAP in rat scalp.

(A–C) Double labelling of newborn (2–3-day-old) rat scalp. (A) staining for GS (green, pAb), (B) staining for GFAP (red, mAb), (C) merge. (D–F) Double staining of adult rat scalp sections. (D) staining for GS, (E) staining for GFAP, (F) merge. The insert in (C) shows single keratinocytes strongly stained for GS. White arrowheads in the insert point to GS-positive cells migrating towards the outer layer of the skin. The insert in (F) shows the prominent differences in the ratio of GS and GFAP in neighboring regions of the epidermis of adult rat skin evidenced by green, yellow or orange staining. Red arrows in (A–F) point to stratum basale, white arrows to stratum spinosum, magenta arrows to stratum granulosum, yellow arrows to stratum corneum. (G–L) Co-expression of GS and GFAP in adult rat leg skin. (G–I) Double labelling of skin section from the lateral surface of adult rat leg skin. (G) staining for GS (green, pAb), (H) staining for GFAP (red, mAb), (I) merge. (J–L) Double staining of skin section from the medial surface of adult rat leg skin. (J) staining for GS, (K) staining for GFAP, (L) merge. White arrows in (G–L) point to the portion of the skin in which GS can be barely detected, whereas red arrows point to the portion of the skin in which GS is strongly expressed. Binding of specific antibodies was visualized with FITC-conjugated goat anti-rabbit IgG (A, D, G, J) and Cy3-conjugated goat anti-mouse IgG (B, E, H, K). Scale bar: (A–C) 10 μm; (D–F) 20 μm.

Figure 5. Co-expression of GS and metallothionein in rat skin sections and in cultured human keratinocytes and rat brain astrocytes.

(A, B) Double-staining for GS (green) and MT (red) in newborn ratscalp sections. (B) Expression of GS (green) and MT (red) in rat scalp hair follicle. Cell nuclei in A–L are stained with DAPI (blue). (C) Co-expression of GS (green) and MT (red) by cultured newborn rat astroglial cells. (D–F) Expression of GS (D) and MT (E) and their co-localization (F, merge) in cultured human skin keratinocytes. GS was detected using rabbit antiserum and FITC-conjugated goat anti-rabbit IgG, whereas MT was detected using mAb MT and Cy3-conjugated goat anti-mouse IgG. Blue arrows point to stratum corneum, yellow arrows in (A) to stratum basale, magenta arrows to the keratinocytes forming the stratum granulosum, red arrows in A and B to the hair channel, green arrows to the outer sheet of the hair follicle, white arrows to fibroblasts; yellow arrowheads in (B) to sweet glands. Red arrows in (D–F) point to cells which predominantly express MT, green arrows in (D–F) point to cells which predominantly express GS. Scale bar: in (A, D, E, F) 10 μm; in (B, C) 20 μm.

Figure 6. Regulatory features of GS in HaCaT cells.

(A) Effect of Gln on the specific activity of GS. Cells were cultivated over 96 h with (squares) and w/o (circles) 5 mM Gln. Medium was renewed every day. Each point represents the mean±SD of 3 determinations for a representative culture. (B) Effect of cultivation time on the specific activity of GS. Cells were cultivated for the times indicated and then harvested. Culture medium was exchanged 3 times a week. Each point represents the mean±SD of 3 determinations for a representative culture. (C) Effect of dexamethasone on the specific activity of GS. Cells were incubated with the indicated concentrations of dexamethasone for 24 h. Each point represents the mean±SD of 3 determinations. (D) Effect of ammonium chloride on the specific activity of GS. Cells were incubated for 24 h in the presence of the indicated concentrations of NH₄Cl. Each point represents the mean±SD of 3 determinations for a representative culture. Statistically significant from controls: *, p<0.05; **, p<0.01; ***, p<0.001.