A mutation in the mucosal keratin K4 is associated with oral white sponge nevus.

White sponge nevus (WSN) is a benign autosomal dominant disorder which affects non-cornifying stratified squamous epithelia (MIM 193900) (ref. 1). Phenotypically it presents as white 'spongy' plaques (oral leukokeratoses), most commonly in the mouth but also reported in the esophagus and anogenital mucosa. Histologically, the plaques show evidence of hyperproliferation, acanthosis and tonofilament aggregation. These types of pathogenic changes are characteristic of many of the epidermal keratin disorders. Keratins are expressed in pairs by epithelial cells in a tissue and cell specific manner. The major differentiation specific keratins of the buccal mucosa, nasal, esophageal and anogenital epithelia are K4 and K13 (ref. 7). The tissue distribution and nature of the lesions in patients affected by WSN suggested that mutations in K4 and/or K13 might be responsible for this disorder. We have now confirmed this hypothesis and report here a three base-pair (bp) deletion in the helix initiation peptide of K4 in affected members from two families with this condition.

Missing links: Weber-Cockayne keratin mutations implicate the L12 linker domain in effective cytoskeleton function.

We have identified mutations in keratins K5 (Arg331Cys) and K14 (Val270Met) in two kinships affected by the dominantly-inherited skin blistering disease, Weber-Cockayne epidermolysis bullosa simplex (EBS-WC). Linkage analysis, DNA sequencing and clinical and ultrastructural analysis are combined to provide the first detailed description of classical EBS-WC. Both phenotypes show similar blistering on trauma, indicating that both mutations compromise the structural resilience of the basal keratinocytes by affecting the keratin cytoskeleton. The location of these mutations in the L12 linker, which bisects the alpha-helical rod region of intermediate filament proteins, identifies another keratin mutation cluster leading to hereditary skin fragility syndromes.

Keratin 16 and keratin 17 mutations cause pachyonychia congenita.

Pachyonychia congenita (PC) is a group of autosomal dominant disorders characterized by dystrophic nails and other ectodermal aberrations. A gene for Jackson-Lawler PC was recently mapped to the type I keratin cluster on 17q. Here, we show that a heterozygous missense mutation in the helix initiation motif of K17 (Asn92Asp) co-segregates with the disease in this kindred. We also show that Jadassohn-Lewandowsky PC is caused by a heterozygous missense mutation in the helix initiation peptide of K16 (Leu130Pro). The known expression patterns of these keratins in epidermal structures correlates with the specific abnormalities observed in each form of PC.

Plectin deficiency results in muscular dystrophy with epidermolysis bullosa.

We report that mutation in the gene for plectin, a cytoskeleton-membrane anchorage protein, is a cause of autosomal recessive muscular dystrophy associated with skin blistering (epidermolysis bullosa simplex). The evidence comes from absence of plectin by antibody staining in affected individuals from four families, supportive genetic analysis (localization of the human plectin gene to chromosome 8q24.13-qter and evidence for disease segregation with markers in this region) and finally the identification of a homozygous frameshift mutation detected in plectin cDNA. Absence of the large multifunctional cytoskeleton protein plectin can simultaneously account for structural failure in both muscle and skin.

Multiple epidermal connexins are expressed in different keratinocyte subpopulations including connexin 31.

Recent genetic studies have demonstrated the importance of epidermal gap junctions with mutations in four beta-connexins associated with autosomal dominant epidermal disease. One of these disorders, erythrokeratoderma variabilis, is associated with germline mutations in the genes encoding connexins (Cx) Cx31 and Cx30.3. Towards understanding the functional mechanism of Cx31 mutations in epidermal disease, we have developed and characterized a polyclonal antibody raised against human Cx31. Using this antibody to immunostain normal epidermis, Cx31 protein was found to be expressed predominately in the stratum granulosum with a punctate pattern of staining at the plasma membrane. In addition, we used reverse transcriptase polymerase chain reaction and, where reagents were available, immunocytochemistry to investigate which other connexins are expressed in the epidermis. Surprisingly, this analysis revealed that there are at least 10 connexins expressed with an overlapping distribution and localization to distinct keratinocyte subpopulations. These data provide additional evidence for multiple gap junction channel types in the human epidermis. Elucidation of this complexity of channel types with respect to specific permeabilities and function of each wildtype and mutant channel type in epidermal biology will require further investigations.

Missense mutations in keratin 17 cause either pachyonychia congenita type 2 or a phenotype resembling steatocystoma multiplex.

Pachyonychia congenita (PC) is a group of autosomal dominant ectodermal dysplasias in which the main phenotypic characteristic is hypertrophic nail dystrophy. In the Jackson-Lawler form (PC-2), pachyonychia is accompanied by multiple pilosebaceous cysts, natal teeth, and hair abnormalities. By direct sequencing of genomic PCR products, we report heterozygous K17 missense mutations in the same conserved protein motif in a further five PC-2 families (K17 N92S in one familial and three sporadic cases; K17 Y98D in one familial case) confirming that mutations in this gene are a common cause of PC-2. We also show heterozygous missense mutations in K17 (N92H and R94H) in two families diagnosed as steatocystoma multiplex. Mild nail defects were observed in some but not all of these patients on clinical re-evaluation of these families. All the K17 mutations reported here were shown to co-segregate with the disease in the pedigrees analyzed and were excluded from 100 unaffected, unrelated chromosomes by restriction enzyme analysis of K17 genomic PCR products. We conclude that phenotypic variation is observed with K17 mutations, as is the case with other keratin disorders.

Donor splice site mutation in keratin 5 causes in-frame removal of 22 amino acids of H1 and 1A rod domains in Dowling-Meara epidermolysis bullosa simplex.

Epidermolysis bullosa simplex (EBS) arises from mutations within the keratin 5 and 14 (K5 and K14) genes which alter the integrity of basal keratinocytes cytoskeleton. The majority of these defects are missense mutations in the rod domain, whose locations influence the disease severity. We investigated a large family dominantly affected with the Dowling-Meara form of EBS (EBS-DM). Sequencing of amplified and cloned K5 cDNA from cultured keratinocytes revealed a 66 nucleotide deletion in one allele corresponding to the last 22 amino acid residues encoded by exon 1 (Val164 to Lys185). Sequencing of amplified genomic DNA spanning the mutant region revealed a heterozygous G-to-A transition at +1 position of the consensus GT donor splice site of intron 1 of K5. This mutation leads to the use of an exonic GT cryptic donor splice site, located 66 nucleotides upstream from the normal donor splice site of intron 1. The corresponding peptide deletion includes the last five amino acids of the H1 head domain and the first 17 amino acids of the conserved amino terminal end of the 1A rod domain, including the first two heptad repeats and the helix initiation peptide. The shortened polypeptide is expressed in cultured keratinocytes at levels which are comparable to the normal K5 protein. This is the first splice site mutation to be reported as a cause of EBS-DM. Owing to the functional importance of the removed region, our data strongly suggest that shortened keratin polypeptide can impair keratin filament assembly in a dominant manner and causes EBS-DM.

Neuroendocrine and behavioural responses to hyperosmolality in rats with lesions of the lateral hypothalamus made by N-methyl-D-aspartate.

Excitotoxic lesions of rat lateral hypothalamus produce impairments in eating and drinking, but not motor deficits. However, it has not been established what causes these eating and drinking impairments. In the present experiments, drinking, plasma osmolality and arginine-vasopressin concentration were measured in lateral hypothalamic-lesioned and control rats following systemic injection of hypertonic saline. In response to hyperosmolality, N-methyl-D-aspartate-lesioned rats drank significantly less than controls but showed normal increases in plasma osmolality and arginine-vasopressin concentration. This dissociation of neuroendocrine and behavioural responses suggests that the impairment of rats with excitotoxic lesions of the lateral hypothalamus is unrelated to physiological (as opposed to behavioural) mechanisms of homeostasis.

Excitotoxic lesions of the pedunculopontine tegmental nucleus of the rat. I. Comparison of the effects of various excitotoxins, with particular reference to the loss of immunohistochemically identified cholinergic neurons.

The pedunculopontine tegmental nucleus (PPTg) has been shown to have cholinergic connections with the thalamus and basal ganglia. The ability of various doses of the excitotoxins (alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) (AMPA), folate, ibotenate, kainate, N-methyl-D-aspartate (NMDA), quinolinate and quisqualate to make lesions in the PPTg was examined, with particular reference to their ability to destroy cholinergic neurons identified using choline acetyltransferase (ChAT) immunohistochemistry. All of the toxins induced convulsive activity on recovery from surgical anesthesia and all except folate made lesions in the PPTg and surrounding structures. The size of the lesions was computed following examination of Cresyl violet stained sections. The largest lesions were made by kainate = AMPA greater than NMDA = ibotenate greater than quisqualate = quinolinate. All of the toxins destroyed cholinergic neurons, higher doses producing greater loss than lower. The ratio of cholinergic cell loss to general neuronal loss (assessed by Cresyl violet staining) was also computed, revealing marked differences between the toxins. Statistical analysis showed that there were significant differences between excitotoxins in terms of this ratio, but these were accounted for by the low dose of quinolinate (24 nmol) producing a significantly greater ratio of damage (12.18:1) than every other toxin. (Next highest ratio: quisqualate 60 nmol, 6.22:1.) Between the other toxins (kainate, AMPA, ibotenate, quisqualate, NMDA and the high dose of quinolinate) there were no statistically significant differences. Intense calcium deposits (stained by Alizarin red) were found frequently and often defined the borders of the lesion. Tyrosine hydroxylase immunohistochemistry revealed axons running below and into the area of lesioned tissue suggesting strongly that fibers were undamaged by the lesions. We conclude that in the PPTg, different excitotoxins make discriminably different lesions, both quantitatively and qualitatively. Unlike excitotoxic lesions in the basal forebrain quinolinate, not quisqualate, made the most selective lesions of cholinergic neurons and, unlike excitotoxic lesions in the septal nuclei, non-myelinated fibers were spared by ibotenate. The implications of these data for research into brainstem mechanisms of Parkinson's disease are discussed.

Excitotoxic lesions of the pedunculopontine tegmental nucleus of the rat. II. Examination of eating and drinking, rotation, and reaching and grasping following unilateral ibotenate or quinolinate lesions.

The pedunculopontine tegmental nucleus (PPTg) contains a population of cholinergic neurons thought to be part of the ascending reticular activating system, and non-cholinergic neurons. In the previous study it was shown that various excitotoxins made effective lesions of cholinergic neurons in the PPTg but that quinolinate made smaller lesions in the non-cholinergic population, making it more selective than any other excitotoxin. The purpose of the present experiment was, first, to make lesions of cholinergic neurons throughout the length of the PPTg by infusing toxin at two different sites within it; and second, to examine simple motor activities in rats bearing either quinolinate or ibotenate lesions of the PPTg, and contrast these with the deficits seen after 6-hydroxydopamine (6-OHDA) induced lesions of mesostriatal dopamine (DA)-containing neurons. Post-mortem examination was carried out using choline acetyltransferase (ChAT) and tyrosine hydroxylase (TOH) immunohistochemistry, and routine Nissl staining. Both quinolinate and ibotenate destroyed approximately 75% of ChAT-positive neurons in the PPTg, but damage to non-cholinergic neurons (assessed by Nissl staining) was twice as great following ibotenate as quinolinate. 6-OHDA induced almost complete lesions of mesostriatal DA neurons, assessed by TOH immunohistochemistry. DA depleted rats showed deficits in drinking and spilled more food in the first 2 weeks after surgery, and were unable to reach or grasp food pellets in the staircase test. They also showed strong ipsilateral turning in response to amphetamine and contralateral turning to apomorphine. Quinolinate lesioned rats had no eating or drinking impairment in the home cage but showed a reaching (though not grasping) disability in the staircase test. They had a mild ipsilateral bias following amphetamine. Ibotenate lesioned rats, despite having larger lesions than the quinolinate, showed no deficits in eating or drinking in the home cage, or reaching or grasping disabilities in the staircase test. They did have a mild contralateral bias in response to amphetamine. This dissociation of the effects of quinolinate and ibotenate lesions of the PPTg is consistent with the suggestion that the PPTg has two functionally distinct components, and is attributed to the differential lesion of non-cholinergic neurons by the two excitotoxins.

A novel mutation in the keratin 13 gene causing oral white sponge nevus.

White sponge nevus (WSN) is an autosomal-dominantly inherited form of mucosal leukokeratosis. Defects in keratins, proteins that form the stress-bearing cytoskeleton in epithelia, have been shown to cause several epithelial fragility disorders. Recently, mutations in the genes encoding mucosal-specific keratins K4 and K13 were shown to be the underlying cause of WSN. We have studied a large Scottish family with 19 persons affected by WSN in four generations. The K4 locus was excluded by genetic linkage analysis; however, genetic linkage consistent with a K13 defect was obtained. Subsequently, a heterozygous missense mutation 335A>G was detected in exon 1 of the KRT13 gene, predicting the amino acid change N112S in the 1A domain of the K13 polypeptide. The mutation was confirmed in affected family members and was excluded from 50 unaffected people by restriction enzyme analysis. These results confirm that mucosal keratin defects are the cause of WSN.

Pharmacological characteristics of beta-adrenoceptor binding sites in intact and sympathectomized rat spleen.

[3H]-(--)-Dihydroalprenolol ([3H]-DHA) binds to rat spleen membranes with a dissociation equilibrium constant (KD) of about 0.7 nM and a maximal number of binding sites of 272 fmoles x mg protein-1. Approximately 90% of the sites labelled by 1 nM [3H]-DHA possess classical properties of beta-adrenoceptors. The labelled ligand is stereospecifically displaced by the isomers of propranolol and the order of potency of agonists is: isoprenaline greater than adrenaline greater than noradrenaline. Highly selective beta 1 antagonists such as practolol and atenolol displaced [3H]-DHA from spleen membranes in a biphasic manner indicating the co-existence of beta 1 and beta 2 sites (30--35% beta 1; 65--70% beta 2) in this tissue. Support for this classification was provided by the binding of the beta 2 agonist procaterol to spleen membranes. This drug possessed high affinity only for those sites that have low affinity for the beta 1 selective agents. Chemical sympathectomy induced by chronic 6-hydroxydopamine administration did not alter the number or the pharmacological properties of beta-adrenoceptor binding sites. The results suggest that both beta 1 and beta 2 adrenoceptors are probably postsynaptic in spleen.

Control of cultured epithelial (MDCK) cell transport function: identification of a beta-adrenoceptor coupled to adenylate cyclase.

A catecholamine-sensitive adenylate cyclase activity was observed in cell homogenates of cultured renal epithelial (MDCK) cells. 10 microM isoprenaline gave a 2.85-fold increase in adenylate cyclase activity above basal levels. A series of adrenoceptor agonists gave a relative potency series of isoprenaline greater than adrenaline greater than noradrenaline (K1(9) values of 1.9 X 10(-7), 1.6 X 10(-6) and 1.9 X 10(-5) M respectively), consistent with activation of a beta-adrenoceptor. Intracellular accumulation of cyclic AMP was also stimulated by 10 microM isoprenaline, peak values being observed after 2 min, followed by a decline to lower maintained levels. The phosphodiesterase inhibitor isobutylmethylzanthine (1 mM) augmented isoprenaline-stimulated cyclic AMP accumulation. In epithelial preparations of MDCK cells grown upon Millipore filters and mounted in Ussing chambers isoprenaline was only effective in elevating intracellular cyclic AMP contents when applied to the basal cell surfaces. Direct measurement of beta-adrenoceptor density and subtype was determined by (+/-)-3-[125I]iodocyanopindolol binding to MDCK cell homogenates. Binding consisted of a saturable component (Vmax = 14.9 fmol/mg cell protein) of high molar affinity (Kd = 10.8 pM) and a non-saturable component which showed a linear dependence on iodocyanopindolol concentration. In addition to the high-affinity binding site, dissociation kinetics revealed a low-affinity component (Kd = 450 pM) comprising 24% of saturable binding. Competition of (+/-)-3-[125I]iodocyanopindolol binding with beta-adrenoceptor agonists and antagonists was entirely consistent with the existence of a beta 2-adrenoceptor. Examination of various MDCK cultures and clones revealed the existence of MDCK cultures whose adenylate cyclase activity was unresponsive to catecholamine stimulation; this correlated with a reduced or undetectable level of (+/-)-3-[125I]iodocyanopindolol binding. The control of transepithelial chloride transport in MDCK epithelia by catecholamines is discussed.

Degradation of [125I]-atrial natriuretic peptide by a soluble metallopeptidase isolated from rat ventricular myocytes.

Atrial natriuretic peptide is rapidly degraded by a soluble, heat labile peptidase isolated from ventricular myocytes. Degradation of [125I]-ANP is antagonized by unlabelled ANP, bradykinin, glucagon, 1,10-phenanthroline, PCMB, EDTA and the bacterial antibiotic bacitracin, but not by phenylmethylsulphonyl fluoride, aprotinin, phosphoramidon, E-64, amastatin or the ACE inhibitor SQ 20881 and bradykinin potentiator C. In addition neither bovine serum albumin nor caesin afforded any protection against degradation. Peptidase activity was optimal at pH values above 8.5. The peptidase is likely to be of intracellular origin and may contribute to the extensive ANP degradative activity found in various ventricular muscle preparations.

Cholinergic stimulation of substantia nigra: abolition of carbachol-induced eating by unilateral 6-hydroxydopamine lesion of nigrostriatal dopamine neurones.

Microinjection of cholinergic agonists into the substantia nigra is known to elicit increases in eating, drinking and sexual behaviour under appropriate circumstances. It has been suggested that these effects are dependent on stimulation of nigrostriatal dopamine-containing neurones in the substantia nigra pars compacta, but no direct evidence has confirmed this. The present experiment was therefore undertaken to determine whether unilateral lesions of nigrostriatal dopamine neurones made by 6-hydroxydopamine would attenuate or abolish eating in satiated rats elicited by intranigral microinjection of the muscarinic agonist carbachol. Two groups of rats were tested: a 6-hydroxydopamine- and a sham-lesion group. Before lesions were made intranigral microinjection of 0.5 microgram/0.5 microliter carbachol stimulated significantly more eating than control microinjections in both groups. After 6-hydroxydopamine lesions, microinjection of carbachol elicited no more eating than vehicle alone. Rats given sham lesions (ascorbate-saline vehicle only) showed increased feeding to intra-nigral carbachol before and after sham-lesioning. Post-mortem analysis by HPLC was used to determine the concentration of dopamine, DOPAC, HVA, serotonin and 5-HIAA in the lesioned and non-lesioned hemispheres of both 6-hydroxydopamine- and sham-lesioned rats. In caudate-putamen there were significant reductions in the concentration of DA (to 50.03% of the level in control sides), DOPAC (to 49.34%) and HVA (to 63.98%) in the 6-hydroxydopamine-lesioned but not sham-lesioned rats. The concentration of dopamine, DOPAC and HVA were not affected in the nucleus accumbens. The turnover of dopamine (assessed by calculating the ratio of dopamine to DOPAC) in the caudate-putamen but not nucleus accumbens was also altered by the 6-hydroxydopamine lesions.(ABSTRACT TRUNCATED AT 250 WORDS)

An investigation of [3H]bumetanide uptake in a cultured renal cell line (MDCK).

The inhibition of ouabain-insensitive 86Rb+ (K+) flux in a cultured renal cell line (MDCK) by a series of loop diuretics, including bumetanide, (the 'cotransport' flux component) has been determined in order to define the concentration range over which [3H]bumetanide would be expected to bind to the membrane transporter involved. Half-maximal inhibition by bumetanide was observed at 0.26 +/- 0.12 (S.D.) microM. The time and concentration dependence of [3H]bumetanide uptake in intact MDCK cells has been determined in experimental conditions, which were as far as possible, identical to inhibition of K+ flux. Total cellular uptake of [3H]bumetanide (0-1 microM) may be separated into a linear component and a component displaying sigmoidal saturation kinetics with a concentration giving half-maximal uptake of 0.33 +/- 0.17 (S.D.) microM, maximal uptake of 1.17 +/- 0.47 pmol/10(6) cells, and a Hill coefficient of 1.59 +/- 0.28. There is also evidence for a second component of [3H]bumetanide uptake of lower affinity (less than 5 microM). Competition of [3H]bumetanide uptake by a series of loop diuretics at varying concentrations gives an order of potency identical to that observed for inhibition of the ouabain-insensitive 86Rb+ influx. The magnitude of the saturable component of [3H]bumetanide uptake is correlated with the magnitude of the diuretic-sensitive 86Rb+ influx in MDCK cells and in a variety of other cultured cells. The relationship between the diuretic-sensitive transport, the saturable component of [3H]bumetanide uptake and the cellular location of bumetanide uptake is discussed.

Loop-diuretic inhibition of adrenaline-stimulated Cl- secretion in a cultured epithelium of renal origin (MDCK).

Loop diuretics (furosemide, bumetanide, piretanide) inhibit the adrenaline-stimulated short-circuit current due to transepithelial Cl- secretion in cultured renal epithelial layers (MDCK). The inhibition of Cl- secretion by loop diuretics is consistent with the presence of basal-lateral 'cotransport' since inhibition is observed only with the basal applications of loop diuretics, is of high potency (half-maximal bumetanide inhibition being observed at 0.8 microM, bumetanide being more potent than furosemide) and is without effect upon the adrenaline-stimulated increase in tissue conductance. Loop diuretics are also shown to inhibit a component of K+ efflux across the basal-lateral surfaces. Cellular uptake of [3H]bumetanide across both apical and basal surfaces of intact epithelial layers was measured in order to localize the cotransport system. A component of cellular [3H]bumetanide uptake sensitive to competition by 0.1 mM unlabelled loop diuretic is only observed from the basal-lateral cell surfaces. There is no evidence for transepithelial bumetanide secretion as is seen in renal cortex.