Functional improvement of mutant keratin cells on addition of desmin: an alternative approach to gene therapy for dominant diseases.

A major challenge to the concept of gene therapy for dominant disorders is the silencing or repairing of the mutant allele. Supplementation therapy is an alternative approach that aims to bypass the defective gene by inducing the expression of another gene, with similar function but not susceptible to the disrupting effect of the mutant one. Epidermolysis bullosa simplex (EBS) is a genetic skin fragility disorder caused by mutations in the genes for keratins K5 or K14, the intermediate filaments present in the basal cells of the epidermis. Keratin diseases are nearly all dominant in their inheritance. In cultured keratinocytes, mutant keratin renders cells more sensitive to a variety of stress stimuli such as osmotic shock, heat shock or scratch wounding. Using a 'severe' disease cell culture model system, we demonstrate reversion towards wild-type responses to stress after transfection with human desmin, an intermediate filament protein normally expressed in muscle cells. Such a supplementation therapy approach could be widely applicable to patients with related individual mutations and would avoid some of the financial obstacles to gene therapy for rare diseases.

K15 expression implies lateral differentiation within stratified epithelial basal cells.

Keratins are intermediate filament proteins whose expression in epithelial tissues is closely linked to their differentiated state. The greatest complexity of this expression is seen in the epidermis and associated structures. The critical basal (proliferative) cell layer expresses the major keratin pair, K5 and K14, but it also expresses an additional type I keratin, K15, about which far less is known. We have compared the expression of K15 with K14 in normal, pathological, and tissue culture contexts; distinct differences in their expression patterns have been observed that imply different regulation and function for these two genes. K15 appears to be preferentially expressed in stable or slowly turning over basal cells. In steady-state epidermis, K15 is present in higher amounts in basal cells of thin skin but in lower amounts in the rapidly turning over thick plantar skin. Although remaining high in basal cell carcinomas (noninvasive) it is suppressed in squamous cell carcinomas (which frequently metastasize). Wounding-stimulated epidermis loses K15 expression, whereas K14 is unchanged. In cultured keratinocytes, K15 levels are suppressed until the culture stratifies, whereas K14 is constitutively expressed throughout. Therefore, unlike K14, which appears to be a fundamental component of all keratinocytes, K15 expression appears to be more tightly coupled to a mature basal keratinocyte phenotype.

Characterization of natriuretic peptide receptor subtypes in the AtT-20 pituitary tumour cell line.

Receptors for the natriuretic peptide family have been characterized in the adrenocorticotrophic hormone (ACTH)-secreting AtT-20 pituitary tumour cell line. Northern blot analysis detected mRNA transcripts for the guanylate cyclase-linked GC-B receptor subtype. There was no evidence for the expression of either guanylate cyclase-linked GC-A receptor or atrial natriuretic peptide (ANP)-C (clearance) receptor mRNAs. Cyclic GMP production in AtT-20 cells was stimulated up to 200-fold by C-type natriuretic peptide (CNP), which was 10- and 20 times as effective as equivalent concentrations of brain natriuretic peptide and ANP respectively. Cyclic GMP dose-response curves to CNP failed to show any signs of saturation even at concentrations up to 30 microM, indicating a relatively low affinity of CNP for the GC-B receptor. Although CNP induced large stimulations in cyclic GMP production, specific binding of [125I-Tyr0]CNP could not be demonstrated in AtT-20 cells. The absence of specific binding with this radiolabelled analogue is possibly due to a reduced affinity for the GC-B receptor, as CNP analogues with N-terminal modifications such as [Tyr0]CNP and [127I-Tyr0]CNP exhibited reduced abilities to stimulate cyclic GMP production in these cells. Despite elevating cyclic GMP levels, CNP had no effect on basal or corticotrophin-releasing factor-stimulating ACTH release from the cells. These results show that the guanylate cyclase-coupled GC-B receptor is the only natriuretic peptide receptor subtype expressed in AtT-20 cells. Although CNP can markedly stimulate cyclic GMP production in these cells, there is incomplete expression of the normal natriuretic peptide-induced inhibitory pathway of ACTH secretion at some point distal to the production of cyclic GMP.

Discrepancy between the short and long term effects of ouabain on the sodium pumps of human cells grown in culture.

1. Human cells (HeLa) were cultured for periods up to 48 h in growth medium in the absence or presence of a range of concentrations of cardiac glycosides. In some experiments the potassium concentration of the medium was varied between 0.3 mM and the usual 5 mM. 2. For periods up to 2 h in ouabain the association and dissociation rate constants were measured and the equilibrium binding constant (KD) calculated; the apparent equilibrium binding constant (K'D) was measured after 1-2 days growth in ouabain. 3. Ouabain had a K'D after 2 days of 2-6 nM in 5 mM K+ growth medium, a 4 fold greater blocking effect on sodium pumps after 2 days than expected from the association and dissociation rate constants measured in untreated or previously ouabain-treated cells. 4. This effect was: (a) approximately the same over a range of external potassium concentrations from 0.3 to 5 mM, although the absolute effect of ouabain over this range of potassium was much different; (b) probably not due to different isoforms of pumps in cells grown in ouabain compared to untreated cells; (c) apparently not a consequence of internalisation of pump-glycoside complexes. 5. We conclude that ouabain has only a limited access to sodium pumps in whole cells; this could be because sodium pumps cycle continuously through an inaccessible region of the plasma membrane. This effect needs to be considered both in the assessment of the magnitude of the long term effects of cardiac glycosides on cells, and in the measurement of the glycoside affinities of various isoforms of the pump.

The effects of monensin on the abundance of mRNA(alpha) and of sodium pumps in human cultured cells.

HeLa or MRC5-VI cells were grown for up to 1 day in media containing monensin at concentrations up to 10 microM. We measured the sodium pump density of the plasma membrane with [3H]ouabain and the mRNA for the alpha-subunit of the pump by hybridization to a cDNA probe. The sodium and potassium concentrations were measured under similar conditions, and in some experiments the rate of internalization of the sodium pumps estimated by using [3H]ouabain uptake into the cell. We found that the relationship between sodium pump density and [Na+]i was well described by Michaelis-Menten kinetics with a Km of 12 mM-[Na+]i and a Vmax twice the normal value. There was no obvious relationship between cell potassium and pump density. The relationship between sodium pump density and [Na+]i was the same as that found by growing the cells in low-potassium medium, so we conclude that the manner of raising [Na+]i is not important, merely the final value. In conditions of raised intracellular sodium there was an increase in the mRNA for the alpha-subunit of the pump, but there was no slowing of the rate of internalization of the pumps from the plasma membrane. We conclude that the increased density of pumps is due to an increased synthesis rather than a decreased internalization rate of pumps, suggesting that the cell can control the method of upregulation.

Transient changes in permeability in HeLa and L cells during detachment from a substrate.

The Na+ and K+ gradients of HeLa cells approach that of the medium during removal from a substrate with trypsin, but then recover during the next 15 min. The recovery is blocked by ouabain or the cold, but is unaffected by bumetanide. The effect is also obtained in cells which have no intercellular connexions, and in cells whose interior is made acid with CO2. Removal of cells with EDTA, pronase E and dispase has similar effects. It does not occur, or is greatly reduced, in cells already rounded up. Substances of molecular weight up to 5000 (e.g. inulin) also cross the cell membrane during this phase. We think that the effect is due to a transient increase in leakiness of the cell during rounding up, possibly due to the detachment of the 'feet' holding the cells onto the substrate. The transient increase in permeability of these cells may be a valuable method of introducing large molecules into them.