The programmed cell death of the stratified squamous epithelial cells comprising human epidermis culminates in abrupt transition of viable granular keratinocytes (KC) into dead corneocytes sloughed by the skin. The granular cell-corneocyte transition is associated with a loss in volume and dry cell weight but the mechanism for and biological significance of this form of keratinocyte apoptosis remain obscure. We show that terminally differentiated KC extrude into the intercellular spaces of living epidermis the cytoplasmic buds containing randomly congregated components of the cytosol as well as filaggrin, a precursor of the natural moisturizing factor. The discharge of secretory product is reminiscent of holocrine secretion, suggesting the term ‘apoptotic secretion’ for this novel, essential step in the process of cornification. The secretory product may become a part of the glycocalyx (a.k.a. ‘intercellular cement substance’ of epidermis) and serve as a humectant that counterbalances the osmotic pressure imposed by the natural moisturizing factor located in the stratum corneum comprised by corneocytes. The apoptotic secretion commences upon secretagouge action of acetylcholine which is synthesized and released by KC. A combination of a cholinergic nicotinic agonist and a muscarinic antagonist which increases intracellular calcium levels is required to trigger the apoptotic secretion. Analysis of the relative amounts of cholinergic enzymes and receptors expressed by KC capable of secretion and the pharmacological profiles of secretion regulation revealed an upward concentration gradient of free acetylcholine in epidermis which may provide for its unopposed secretagogue action via the m1 muscarinic and the α7, and α9 nicotinic receptor types expressed by KC at the latest stage of their development in the epidermis.

Both the delicate texture of the skin and its impermeability to environmental hazards depend upon the ability of the superficial layer of the skin, the epidermis, to preserve water through intercellular occlusion and humectancy (reviewed by Elias, 1996; Rawlings et al., 1994). In human epidermis, the stratified epithelial cells, keratinocytes (KC), undergo metamorphosis from immature, small basal cells into mature, large granular KC through a series of genetically determined steps of differentiation culminating in programmed cell death, or cornification. The transition of granular KC into flattened corneocytes is abrupt and associated with a loss in volume and dry cell weight in a range from 45 to 86% (Meyer et al., 1970). Biological significance of cornification stems from the ability of the stratum corneum to protect the body from water loss and external chemical injury. Although it is well known that synthetic products of granular KC subserve barrier function of the stratum corneum, forming a ‘mortar’ in a ‘bricks and mortar’ model of human stratum corneum (Menon et al., 1986), and that calcium-dependent enzymes crosslink intracellular proteins that envelop a corneocyte, a ‘brick,’ the physiological mechanism for and biological significance of rapid volume loss at the final stage of the programmed cell death in epidermis remain unknown.

In the cytosol of terminally differentiated KC, the humectant precursor protein profilaggrin is associated with optically dense keratohyaline granules, and the barrier lipids are contained within lamellar bodies. The discharge of lamellar bodies by terminally differentiated KC is well characterized (Odland and Holbrook, 1987), and it has been recently proposed to name the outermost granular cell a ‘specialized secretory cell,’ to encompass the specialized features of granular KC (Elias et al., 1998). When granular KC make their transition into anucleated corneocytes, the dissolution of keratohyaline granules coincides with enzymatic processing of profilaggrin into mature filaggrin (reviewed by Dale et al., 1994). Transitional cells become almost filled by homogenous keratohyaline masses (Odland, 1991), and filaggrin immunoreactivity is detectable mainly in the cytoplasm (Ishida-Yamamoto et al., 1998). Careful analysis of published series of electron microphotographs depicting the granular cell-corneocyte transition (Odland, 1971; Odland, 1991) reveals that in addition to extruding lamellar bodies, the outermost granular cells exhibit budding of the plasma membrane. The cytoplasmic buds contain keratohyaline material which seems to be extruded by the cells, suggesting that terminally differentiated KC secrete filaggrin.

Exocrine secretion is common for different types of epithelial cells that reside in the mucocutaneous tissues and use acetylcholine (ACh) as a secretagouge for their secretion (Ikeda et al., 1995; Puchelle et al., 1991; Soltoff and Toker, 1995). The epithelium lining aerodigestive tracts and comprising human epidermis contains high amounts of free ACh (Grando et al., 1993b; Klapproth et al., 1997). In these locations, nonneuronal ACh exhibits a plethora of biological effects, including regulation of cell viability, proliferation, motility, adhesion and differentiation (reviewed by Grando, 1997; Sastry and Sadavongvivad, 1978; Wessler et al., 1999). ACh exerts its physiological control of human KC via both the nicotinic and the muscarinic pathways, because both classes of ACh receptors are simultaneously expressed by these cells (Grando et al., 1995a; Grando et al., 1995b). Direct pharmacologic action of the ACh congener nicotine on KC results in accelerated differentiation (senescence), characterized by increased filaggrin production, followed by cell death, or cornification (Grando et al., 1996a; Kwon et al., 1999; Theilig et al., 1994). We hypothesized that: (1) programmed cell death in epidermis includes a step when terminally differentiated KC extrude pieces of their cytoplasm containing keratohyaline material; and (2) this keratinocyte secretion is controlled by ACh.

In this study, the ultrastructural examination of terminally differentiated KC in human skin and cultured monolayers showed spontaneous extrusion of the filaggrin-positive cytoplasmic buds containing various organelles. This keratinocyte secretion was mediated by ACh, acting through its m1 muscarinic, and α7 and α9 nicotinic receptors that use calcium as a second messenger. Since the concentration of free ACh is determined by the ratios of its synthesis by choline acetyltransferase (ChAT) and hydrolysis by acetylcholinesterase (AChE), we quantified relative amounts of both enzymes and free ACh throughout epidermal layers. We found an upward concentration gradient of free ACh in human epidermis which may provide for its unopposed secretagouge action exclusively in the stratum granulosum. We conclude that programmed cell death of epidermal KC culminates in ACh-induced sporadic extrusion of cell components which allows to accomplish the condensation of KC during the granular cell-corneocyte transition, and also releases humectants into the intercellular space of the epidermis.

Cell culture experiments

Keratinocyte cultures were started from normal human neonatal foreskins and grown in 75 cm2 flasks (Corning Glass Works, Coring, NY) in serum-free keratinocyte growth medium containing 0.09 mM Ca2+ (Gibco BRL, Gaithersburg, MD) at 37°C in a humid, 5% CO2 incubator, as we described in detail earlier (Grando et al., 1993a). The purity of cultures was investigated immunocytochemically using DAKO-CK monoclonal mouse antihuman cytokeratin antibody (MNF 116, DAKO Corp., Carpinteria, CA), and was consistently >95%. The viability of the cultured KC, as revealed by the trypan blue dye exclusion test, was consistently >95%. To study morphologic changes, KC were seeded into 6-well tissue culture plates (Falcon 3046; Becton Dickinson Labware, Lincoln Park, NJ) at a cell density of 1×105/well and grown to confluence in 2 ml of growth medium containing 0.09 mM Ca2+. To achieve complete differentiation, some monolayers were incubated for additional 6, 12, 24, 36 or 48 hours in the growth medium supplemented to contain 1.8 mM Ca2+. To observe morphologic changes occurring naturally with terminally differentiated KC in vitro, the monolayers that had reached confluence at 0.09 mM Ca2+ were kept in the culture for additional 2-3 weeks at the same concentration of extracellular Ca2+, which allows slow stratification and crown formation but avoids massive cornification of KC induced by elevation of the concentration of Ca2+ in the growth medium (Grando et al., 1996b). On the day of experiment, the plates were placed on the preheated (37°C) stage of an inverted, computer-linked phase-contrast microscope (Axiovert 135; Carl Zeiss Inc., Thornwood, NY) equipped with a CCD video camera (Photon Technology International, Monmouth Junction, NJ) and observed before and after addition of various concentrations and combinations (see Results) of test cholinergic compounds: atropine, carbachol, cytisine, bromoacetylcholine, mecamylamine, methacholine, muscarine, nicotine, oxotremorine, tropicamide, tubocurarine (all from Sigma Chemical Co., St Louis, MO), propylbenzilylcholine mustard (NEN, Boston, MA) and κ-bungarotoxin (Biotoxins Inc., St Cloud, FL), dissolved in culture medium containing 0.09 mM Ca2+. The pH of the medium containing test compounds was maintained within the range 7.2-7.4, and the osmolarity was 290-310 mOsm/kg under all experimental conditions. The monolayers were videorecorded throughout experiments. The morphologic changes were quantitated by comparing the area (in pixels) covered by each cell prior to (taken as 100%) and after discharge of the cytoplasmic material. Phase-contrast microscopic images were captured on a Power Macintosh 8500/120 computer and analyzed with image analysis software (Scanalytics, Fairfax, VA).

DNA fragmentation assay

To determine the fate of keratinocytes after secretion, a monolayer of the cells that had discharged their cytoplasmic buds in response to a combination of cholinergic drugs (Table 1) was washed with growth medium containing 0.09 mM Ca2+, to remove cholinergic drugs and extruded material, fed with fresh growth medium, incubated for 24 hours in a 5% CO2 and then used in a standard DNA fragmentation assay. Briefly, after washing with prewarmed (37°C) phosphate-buffered saline, the KC were scraped with a rubber policemen, separated by centrifugation and incubated for 3 hours at 55°C in the DNA extraction buffer consisting of 100 mM Tris-HCl (pH 8.0), 200 mM NaCl, 5 mM EDTA, 0.2% sodium dodecyl sulfate (SDS) and 0.2 mg/ml proteinase K. The total cellular DNA was obtained by the addition of isopropanol and subsequent centrifugation at 18,000 g for 1 hour. The pellet was washed with 75% ethanol, dried, and resolved by electrophoresis in a 1.5% agarose gel. The cellular DNA extracted from the intact keratinocyte culture that was established from the same donor and grown in parallel with experimental cells was used as a negative control.

Table 1.

Cholinergic pharmacology of keratinocyte secretion

Cholinergic pharmacology of keratinocyte secretion
Cholinergic pharmacology of keratinocyte secretion

Immunofluorescence assays

The indirect immunofluorescence experiments were performed as detailed previously (Ndoye et al., 1998), using as a substrate cryostat sections of normal human skin or cultured keratinocyte monolayers grown on coverslips. The skin samples obtained from cosmetic or reparative surgical procedures were transported to the laboratory in a fixative containing 25 mM potassium citrate, 5 mM MgSO4, and 5 mM n-ethylmaleimide in distilled water, snapped frozen in liquid nitrogen and stored at −80°C. First passage human foreskin KC were seeded onto glass coverslips, and incubated for two days to form monolayers. To visualize the cholinergic enzymes located intracellularly, skin sections were permeabilized with 100% acetone for 10 minutes. To visualize the cholinergic molecules expressed on the cell surface of KC, the skin specimens and cell monolayers were fixed for 3 minutes in 3% freshly depolymerized paraformaldehyde that contained 7% sucrose, thus avoiding cell permeabilization. Following the protocol described in detail previously (Grando et al., 1995a), after extensive washes in phosphate-buffered saline, the fixed specimens were incubated overnight at 4°C with a primary antibody to ACh (Biodesign International, Kennebunk, ME; dilution 1:1000), AChE (Affinity Bioreagents Inc., Golden, CO; 1:50), ChAT (Chemicon International, Temecula, CA; 1:50), m1 and m4 muscarinic (both diluted 1: 1000) or α7 nicotinic receptors (1:200), characterized by us in the past (Ndoye et al., 1998; Nguyen et al., 2000a; Zia et al., 2000) and currently available from Research & Diagnostic Antibodies (Richmond, CA), or the α9 nicotinic subunit antibody (1:200) recently raised in our laboratory (Nguyen et al., 2000b). Filaggrin was visualized in cell monolayers and horizontal sections of the epidermis using mouse IgG1 monoclonal anti-filaggrin antibody (Argene Inc., North Massapequa, NY; 1:150). To characterize keratinocyte phenotype, in addition to anti-filaggrin antibody, we also used antibodies to the following differentiation and cell cycle markers: Ki 67 antigen (DAKO A/S, Denmark; 1:200), cytokeratins 1/10/11 (Sigma Chemical Co.; 1:100) and 5/6 (Boehringer Mannheim Corp, Indianapolis, IN; 1:100), loricrin (BABCO, Richmond, CA; 1:500), and involucrin (prediluted) and keratinocyte transglutaminase type I (1:10) (both from Biomedical Technologies, Inc. Stoughton, MA). Binding of primary antibody was detected by staining with the appropriate secondary, fluorescein-isothiocyanate-conjugated antibody purchased from Pierce (Rockford, IL) at room temperature for 1 hour. The specificity of antibody binding in the indirect immunofluorescence experiments was demonstrated by abolishing the staining by omitting the primary antibody, by preincubating an antiserum with the specific peptide used for immunization, and by replacing primary antibody with an irrelevant antibody of the same isotype and species as the primary antibody. In particular, mouse IgG1 anti-melanocyte monoclonal antibody Mel.2 (Biogenesis, Sandown, NH; 1:40) was used to control for specific staining by anti-filaggrin antibody. The specimens were examined using Axiovert 135 fluorescence microscope. To calculate relative amounts of cholinergic molecules throughout the epidermis, the acquired immunofluorescent images were analyzed by a semi-quantitative immunofluorescence assay (Ndoye et al., 1998). For each skin specimen, at least three different randomly selected segments in at least three different microscopic fields were analyzed. The results of semi-quantitative immunofluorescence assays were expressed as mean ± s.d. Significance was determined using Student’s t test.

Electron microscopic assays

For transmission electron microscopic observations of terminally differentiated KC, normal human skin specimens were fixed in 2.5% glutaraldehyde overnight followed by exposure to 1% buffered osmium tetroxide, dehydrated in a series of ethanol baths, embedded in SPURR (Electron Microscopic Science, Fort Washington, PA), sectioned to a thickness of 70-90 nm with an LKB Ultramicrotome, picked up on Formvar coated nickel grids, and examined using the EMU4 electron microscope (RCA, Camden, NJ), essentially as described previously (Grando et al., 1995b).

Scanning electron microscopy was used to obtain the three-dimensional images of the cytoplasmic buds extruded by KC in the monolayers exposed to cholinergic drugs. Small amounts of chasing medium, 50-100 μl, containing the extruded spheres were collected promptly after their discharge and placed upon polylysine coated glass slides. The slides were fixed with 2.5% glutaraldehyde in 0.1 M Na-cacodylate buffer for three hours, rinsed 5 times in 0.1 M Na-cacodylate buffer for 5 minutes each followed by fixation in 1% OsO4 for 30 minutes with gentle agitation and a rinse in 0.1 M Na-cacodylate buffer. The slides were next dehydrated in graded ethanol for 5 minutes each (50%, 70%, 80%, 2× 95%, 2× 100%), transported in fresh 100% ethanol to the Samdri-780A critical point dryer (Tousimis Research Corporation, Rockville, MD) and dried at 100 atm, 31.1°C and the CO2 density of 0.48 g/cm3. The slides were then transferred to a platinum sputter coater and coated with a 1 nm layer of elemental platinum. The specimens were mounted on brass chips with carbon tape, grounded with silver epoxy, and observed with the Hitachi S-900 Field Emission Scanning Electron Microscope with an accelerating voltage of 3.5 kV and a flashing voltage of 5.1 kV.

Immunoblotting assays

The presence of filaggrin in the secretory products of KC was determined by western blot analysis of SDS-polyacrylamide gel electrophoresis (SDS-PAGE)-separated proteins, which was performed in accordance to a standard protocol (Nguyen et al., 1998) with minor modifications. Briefly, 20 μg of the lyophilized samples of chasing medium from the keratinocyte monolayers exposed to cholinergic drugs (experiment) or non-treated monolayer (negative control), and of a homogenate of normal human cultured KC (positive control) were loaded into each lane of a 15% SDS-PAGE gels, separated by electrophoresis and electroblotted onto the Immobilon-P membrane (Millipore Corporation, Bedford, MA). The membranes were blocked with 5% milk in Tris-buffered saline for 1 hour at 37°C and exposed for 1 hour at room temperature to a primary mouse anti-filaggrin monoclonal antibody. Binding of primary antibody was visualized using horseradish peroxidase-conjugated goat anti-mouse IgG antibody (Bio-Rad, Hercules, CA), developed with 4-chloro-1-naphthol substrate (Opti-4CN; Bio-Rad). The specificity of staining was determined by the absence of immunoblotting membrane staining when the primary antibody was omitted.

Total RNA extraction and reverse transcription polymerase chain reactions (RT-PCR)

A modification of the Chomczynski and Sacchi (Chomczynski and Sacchi, 1987) method using tri-reagent (Gibco BRL) was used to extract total RNA from cultured KC. Contaminating DNA was removed from the total RNA extraction by incubation for 30 minutes at 37°C in a solution containing 1× DNase I buffer and an RNAse inhibitor, as described previously (Nguyen et al., 2000a). The reaction was stopped by the addition of 2 mM EDTA, and the DNase I inactivated by heating the samples for 10 minutes at 65°C. In order to synthesize first strand cDNA, the DNase treated RNA samples (5 μg) underwent reverse transcription in 1× Superscript II buffer (Gibco BRL), 5 mM dithiothreitol, 60 U RNase Block (Stratagene, La Jolla, CA), 10 μM random decamer primers (Gibco BRL), 600 U Superscript II reverse transcriptase (Gibco BRL), and 1 mM each of dATP, dCTP, dGTP and dTTP, in a final volume of 60 μL. The solution was incubated for 60 minutes at 42°C. The PCR amplification was performed using a protocol described previously (Hall et al., 1998) in a final volume of 50 μl containing the RT product, 1× PCR buffer (Gibco BRL), 0.2 mM each of dATP, dCTP, dGTP, dTTP, 2 U Taq DNA polymerase (Gibco BRL) and 1 μM each of the sense and corresponding antisense primer pairs. Primers for PCR amplification of AChE (sense: 5′-cagcgactgatgcgatactg-3′, antisense: 5′-ctgcttgct- gtagtggtcga-3′; expected product size 312 bp) and ChAT (sense: 5′-gaaatgctccccggaaattc-3′, antisense: 5′-ctcacaaaagccagtgcctc-3′; 291 bp) were designed to cross intron boundaries permitting the differentiation of RNA amplified products from contaminating DNA amplified products. The reaction mixture was heated to 95°C for 5 minutes prior to adding 2 U of Taq DNA polymerase. This extended denaturation was followed by annealing, at the primer specific temperature, for 1 minute and elongation for 2 minutes at 72°C followed by 34 cycles in which denaturation was carried out at 95°C for 1 minute, annealing for 1 minute, and elongation at 72°C for 2 minutes. In the final cycle, the elongation step was increased to 10 minutes. The specificity of each product was determined by comparing the product size, estimated with the use of a 100 bp or 250 bp DNA ladder standard, with the expected size based on the primer design. Further confirmation of the products specificity was obtained by automated DNA sequencing (ABI Prism 377, Perkin Elmer, NJ) following purification of the product from the gel using the silica membrane spin-column technology (QIAquick Spin, Qiagen, Santa Clarita, CA). Each RT-PCR experiment included the following controls: (1) PCR amplification of DNase and non-RT treated total RNA extract to test for amplification artifacts due to contaminating residual genomic DNA; (2) PCR conducted using a DNase- and RT-treated samples in which the extracted RNA template was replaced with sterile water, controlling for contamination of solutions following RNA extraction; and (3) PCR conducted using human DNA as a positive control for the adequacy of the PCP primers used.

Measurement of intracellular free calcium ([Ca2+]i)

The concentration of [Ca2+]i in KC was measured using calcium sensitive Fura-2 acetoxymethyl ester by the fluorescence ratiometric method detailed elsewhere (Zia et al., 2000). Briefly, first passage KC were plated onto two-well glass coverslip chambers (Nunc Nalgene, Naperville, IL) at a density of 1.5×104 cells/cm2, and cultured in a humid CO2 incubator at 37°C. On the day of an experiment, the cells were washed with HEPES/Hanks buffer consisting of 20 mM HEPES, 132 mM NaCl, 5.4 mM KCl, 0.44 mM KH2PO4, 0.34 mM Na2HPO4, 0.41 mM MgSO4, 0.49 mM MgCl2, 0.03 mM CaCl2, 5.5 mM glucose, and 0.05% bovine serum albumin (pH 7.4), loaded for 60 minutes at room temperature with 5 μM Fura-2 (Molecular Probes, Eugene, OR) in the HEPES/Hanks buffer, washed twice with the same buffer, and allowed to recover for 15 minutes by incubating in a CO2 incubator at 37°C. The chamber slide was then mounted on the preheated stage at 37°C of an invert Axiovert 135 microscope and the Fura-2 fluorochrom was excited sequentially at wavelengths of 340 and 380 nm and its emission at 510 nm was detected. The fluorescence was quantified by averaging pixel intensities of twenty cells for each experiment and by subtracting background. At the end of each experiment, 340:380 nm fluorescence ratios were calibrated by measuring minimum and maximum fluorescence.

Spontaneous cytoplasmic budding of terminally differentiated KC in vivo and in vitro

Transmission electron microscopic examination of normal human skin revealed club-shaped buds that projected sporadically from the outermost granular KC (Fig. 1A). The cytoplasmic protrusions were connected to the cells via narrow pedicle (Fig. 1A, long arrows). The endoplasm that filled the protrusion always showed appreciable amounts of stellate masses of keratohyalin, some lysosomal elements and mitochondria (Fig. 1A, short arrows), but could also contain glycogen granules, segments of rough or smooth endoplasmic reticulum and lipid globules (data not shown).

Fig. 1.

Spontaneous extrusion of cytoplasmic buds by terminally differentiated human KC. (A) Transmission electron microscopy of normal human skin at the junction between the stratum corneum and the uppermost layer of living KC. The club-shaped bud projected spontaneously from the granular cell that undergoes apoptosis, as judged from clumping of chromatin in remains of its nucleus. Arrowheads indicate a bilayered leaflet of the cell membrane enveloping the budding cell. Long arrows indicate a narrow plasma membrane pedicle connecting a club-shaped cytoplasmic bud to the cell. The site of connection is designated by an asterisk. Small arrows indicate mitochondria within the nascent cytoplasmic bud. Bar, 3 μm. (B and C) Phase-contrast observations in the aged, i.e. kept for 2 weeks after confluence, culture of normal human KC grown at 0.09 mM extracellular Ca2+. Practically all senescent cells exhibited peripheral cytoplasmic blebbing followed by extrusion of cytoplasmic buds. The underlying keratinocyte monolayer is out of focus because the terminally differentiated KC that secrete spheres are located in organotypic structures crowning the monolayer. The results of immunocytochemical assays showed that KC comprising such organoids are positive for the markers of terminal differentiation filaggrin, loricrin, involucrin, transglutaminase and cytokeratins 1/10/11, and negative for the markers of proliferating cells Ki 67 and cytokeratins 5/6 (not shown). Arrows indicate pedicle connecting a nascent sphere to the cell of its origin (B,C). Arrowheads indicate the spheres that have already pinched off and float free in the culture medium (C). Bars, 25 μm. (D) A phase-contrast image of a balloon-like, floating free sphere which has enlarged in size and lost partially its original optical density. The floating cytoplasmic buds enlarged, burst and disappeared within first 10-15 minutes after detachment from the cells. Bar, 7 μm.

Fig. 1.

Spontaneous extrusion of cytoplasmic buds by terminally differentiated human KC. (A) Transmission electron microscopy of normal human skin at the junction between the stratum corneum and the uppermost layer of living KC. The club-shaped bud projected spontaneously from the granular cell that undergoes apoptosis, as judged from clumping of chromatin in remains of its nucleus. Arrowheads indicate a bilayered leaflet of the cell membrane enveloping the budding cell. Long arrows indicate a narrow plasma membrane pedicle connecting a club-shaped cytoplasmic bud to the cell. The site of connection is designated by an asterisk. Small arrows indicate mitochondria within the nascent cytoplasmic bud. Bar, 3 μm. (B and C) Phase-contrast observations in the aged, i.e. kept for 2 weeks after confluence, culture of normal human KC grown at 0.09 mM extracellular Ca2+. Practically all senescent cells exhibited peripheral cytoplasmic blebbing followed by extrusion of cytoplasmic buds. The underlying keratinocyte monolayer is out of focus because the terminally differentiated KC that secrete spheres are located in organotypic structures crowning the monolayer. The results of immunocytochemical assays showed that KC comprising such organoids are positive for the markers of terminal differentiation filaggrin, loricrin, involucrin, transglutaminase and cytokeratins 1/10/11, and negative for the markers of proliferating cells Ki 67 and cytokeratins 5/6 (not shown). Arrows indicate pedicle connecting a nascent sphere to the cell of its origin (B,C). Arrowheads indicate the spheres that have already pinched off and float free in the culture medium (C). Bars, 25 μm. (D) A phase-contrast image of a balloon-like, floating free sphere which has enlarged in size and lost partially its original optical density. The floating cytoplasmic buds enlarged, burst and disappeared within first 10-15 minutes after detachment from the cells. Bar, 7 μm.

Careful analysis of the images obtained during time-course observations of the organotypic structures crowning the monolayers in aged cultures of normal human epidermal KC also revealed spontaneous extrusion of cytoplasmic buds. The blebbing organoids contained terminally differentiated KC, as judged from their specific staining for the differentiation markers cytokeratins 1/10/11, filaggrin, loricrin, involucrin and transglutaminase, and lack of staining with antibodies to the Ki 67 antigen and the cytokeratins 5/6 (not shown). Peripheral membrane blebbing of terminally differentiated KC culminated in extrusion of pieces of their cytoplasm in the form of optically dense spheres (Fig. 1B,C). Some cytoplasmic knobs were still connected to the cell via narrow pedicle, and some have pinched off and floated free in the culture medium. Time-lapse videorecording demonstrated that within 2-15 minutes after separation from the cell, the spheres grew in size several times, became translucent and reminiscent of large balloons (Fig. 1D), and eventually burst and disappeared.

Visualization of filaggrin in cytoplasmic buds of granular KC using horizontal sections of normal human epidermis

Immunofluorescent staining with anti-filaggrin antibody was used to label keratohyaline material in the cytoplasmic buds extruded by terminally differentiated KC in human skin. The skin samples were sectioned horizontally because standard vertical sectioning does not allow to distinguish cytoplasmic protrusions of granular KC on the background of a larger cell body and/or other cells located in the same epidermal layer. Upon horizontal sectioning, the granular epidermal layer was split into two halves, rendering a monolayer-like appearance to the section. Examination of the sections revealed nascent cytoplasmic protrusions connected to the cells via a clearly distinguishable neck, the cytoplasmic buds that were about to pinch off, yet still connected to the cells via a slender stalk (Fig. 2A), and the buds that had been separated completely from the cells of their origin (Fig. 2B). Both attached and detached buds could be clearly seen in the immunofluorescence images because both contained appreciable amounts of keratohyaline material visualized by anti-filaggrin antibody. The detached buds, compared to the attached ones, stained less brightly, perhaps because they had started to empty their keratohyalin load into the intercellular space.

Fig. 2.

Visualization of filaggrin-positive secretory material extruded spontaneously by terminally differentiated KC in normal human epidermis. (A) Normal human epidermal specimens were snap frozen in liquid nitrogen and horizontally embedded, to allow horizontal sectioning. The series of 4 μm thick horizontal sections were obtained, and treated with mouse monoclonal anti-filaggrin antibody and then with secondary, fluorescein-isothiocyanate-conjugated antibody (see Materials and Methods). The immunostaining at the level of the granular layer reveals the presence of keratohyaline material in a nascent cytoplasmic bud, which is connected to the cell via pedicle (arrow). Bar, 5 μm. (B) The pinched off keratohyaline sphere laying free in the intercellular space of the stratum granulosum. Bar, 7 μm. (C) Negative control showing a lack of fluorescence staining in the specimen of horizontally sectioned epidermis treated with secondary, anti-mouse IgG1 antibody without primary, anti-filaggrin antibody. Bar, 5 μm.

Fig. 2.

Visualization of filaggrin-positive secretory material extruded spontaneously by terminally differentiated KC in normal human epidermis. (A) Normal human epidermal specimens were snap frozen in liquid nitrogen and horizontally embedded, to allow horizontal sectioning. The series of 4 μm thick horizontal sections were obtained, and treated with mouse monoclonal anti-filaggrin antibody and then with secondary, fluorescein-isothiocyanate-conjugated antibody (see Materials and Methods). The immunostaining at the level of the granular layer reveals the presence of keratohyaline material in a nascent cytoplasmic bud, which is connected to the cell via pedicle (arrow). Bar, 5 μm. (B) The pinched off keratohyaline sphere laying free in the intercellular space of the stratum granulosum. Bar, 7 μm. (C) Negative control showing a lack of fluorescence staining in the specimen of horizontally sectioned epidermis treated with secondary, anti-mouse IgG1 antibody without primary, anti-filaggrin antibody. Bar, 5 μm.

No specific staining could be seen in negative control experiments in which the primary anti-filaggrin antibody was either omitted (Fig. 2C) or replaced with an irrelevant, mouse IgG1 anti-melanocyte monoclonal antibody Mel.2 (not shown).

Cholinergic agents trigger keratinocyte secretion in cultures

Confluent monolayers of normal human epidermal KC were grown in culture to confluence at 0.09 mM extracellular Ca2+ and used in phase-contrast experiments testing the ability of cholinergic drugs to elicit cytoplasmic budding. Some monolayers were used in experiments after they had been allowed to differentiate furthermore by either elevating extracellular Ca2+ in the growth medium to 1.8 mM during the last 6, 12, 24, 36, and 48 hours of incubation or extending for additional 2-3 weeks the incubation of confluent monolayers at 0.09 mM Ca2+. Depending on the concentration of extracellular Ca2+ in the growth medium and/or the duration of incubation at the 1.8 mM Ca2+ concentration, the phenotype of KC used in experiments varied from approximately 50 to 100% of cells expressing cytokeratins 1/10/11, filaggrin, loricrin, involucrin and transglutaminase, and from 0 to about 50% of cells expressing the Ki 67 antigen and the cytokeratins 5/6. The 6-well standard tissue and cell culture plates containing keratinocyte monolayers were placed on the preheated (37°C) stage of an inverted phase-contrast microscope, and the cells were fed with fresh, pre-warmed culture medium containing different combinations of test drugs. The changes in cell morphology were recorded using a time-lapse video-recorder.

Since human epidermal KC express both the nicotinic and the muscarinic classes of ACh receptors (Grando et al., 1996a; Grando et al., 1995a; Ndoye et al., 1998; Nguyen et al., 2000b), both of which are stimulated constantly by the ACh that is synthesized and released by KC in culture medium (Grando et al., 1993b), to selectively activate each signaling pathway we had to combine an agonist of one receptor class with an antagonist of the other receptor class. Because cultured KC exhibit high AChE activity (Grando et al., 1993b), we used AChE-resistant agonists. Nicotinic receptors were activated by the reversible agonist nicotine, cytisine or the irreversible agonist bromoacetylcholine (Derkach et al., 1991) and blocked reversibly by mecamylamine and tubocurarine, and irreversibly by κ-bungarotoxin (Grando et al., 1995a). To selectively block the α9-made ACh receptor, we used strychnine (Elgoyhen et al., 1994). Muscarinic receptors were activated by the agonists muscarine, oxotremorine or methacholine and blocked reversibly by atropine or tropicamide and irreversibly by propylbenzilylcholine mustard (Eglen et al., 1994). To simultaneously activate both receptor classes while, at the same, inhibiting reversibly AChE, we used carbachol (DiPalma, 1994). The concentrations of each cholinergic agent used in this study ranged from 1×10−8 to 10−3 M, as suggested by the results of radioligand binding experiments and functional assays measuring cholinergic effects on cultured KC (reviewed by Grando, 1997).

Among the experimental conditions tested, the combinations of drugs providing for an activation of the nicotinic-type ACh receptors at the background of muscarinic receptor inhibition reproducibly elicited extrusion of cytoplasmic buds (Table 1). The discharge could be achieved using the following combinations of cholinergic agonists and antagonists: bromoacetylcholine plus atropine, tropicamide or propylbenzilylcholine mustard, and carbachol plus atropine, tropicamide or propylbenzilylcholine mustard. The detachment of blebs from KC could be abolished by pretreating the monolayers with micromolar concentrations of strychnine, suggesting that activation of α9 nicotinic ACh receptor played especial role in facilitating the release of the secretory product by KC. Nicotine and cytisine did not induce detachment of the cytoplasmic buds from cultured KC, neither were these drugs able to activate α9 nicotinic ACh receptors in a study reported elsewhere (Elgoyhen et al., 1994).

A typical sequence of morphologic events observed in both high Ca2+-induced and naturally differentiated cell monolayers exposed to micromolar concentrations of a nicotinic agonist in the presence of a muscarinic antagonist is shown on Fig. 3. Within 3 to 15 minutes of exposure, depending upon the doses of the test drugs used, optically dense cytoplasmic knobs started to protrude at the periphery of exposed cells (Fig. 3A,B). Over next 5-15 minutes, these small, approximately 1-3 μm in diameter, knobs grew into large, 5-7 μm in diameter, spheres which pinched off and floated away from the cells (Fig. 3C). The spheres could be easily removed from the cultures by collecting chasing medium overlaying the monolayers. The cells in the monolayers exposed to a muscarinic antagonist alone produced reversible cytoplasmic blebbing without discharge of the cytoplasmic buds (not shown). Non-differentiated cells responded to the secretagogue action of cholinergic drugs approximately 2-3 times faster then cells differentiated due to incubation of monolayers, grown at 0.09 mM Ca2+, either at 0.09 mM Ca2+ for 2-3 weeks, or at 1.8 mM Ca2+ for 24-48 hours postconfluence.

Fig. 3.

Time-course observation of extrusion of cytoplasmic buds in response to unopposed nicotinic stimulation observed in a monolayer of normal human KC. (A) The intact monolayer of second passage human foreskin KC which was established in a standard 6-well tissue culture plate and grown to confluence and then incubated for additional 3 weeks at 0.09 mM extracellular Ca2+, as described in Materials and Methods. The immunocytochemical analysis of parallel cultures revealed that such monolayers contain approximately 65% of mature KC that are positive for the markers of terminal differentiation filaggrin, loricrin, involucrin, transglutaminase and cytokeratins 1/10/11, and approximately 35% of immature KC that are positive for the markers of proliferating cells Ki 67 and cytokeratins 5/6. The plate was installed on the preheated (37°C) stage of an inverted microscope and used in experiment. Note: the cells indicated with an asterisk were randomly selected to determine if the discharge of cytoplasmic material leads to cell shrinkage. (B) Same microscopic field 5 minutes after the culture received prewarmed (37°C) medium supplemented to contain 1 mM bromoacetylcholine and 0.1 mM atropine. Note: extensive cytoplasmic budding. (C) Same field 10 minutes after exposure. Note: the buds have grown in size and number, and many spheres have pinched off and float free. (D) Same field after the pinched off spheres were removed by collecting the chasing medium. Note: the cells indicated with an asterisk in A have condensed and shrunk to 91.2±4%, compared to their original size, taken as 100% (n=10). Bar, 50 μm.

Fig. 3.

Time-course observation of extrusion of cytoplasmic buds in response to unopposed nicotinic stimulation observed in a monolayer of normal human KC. (A) The intact monolayer of second passage human foreskin KC which was established in a standard 6-well tissue culture plate and grown to confluence and then incubated for additional 3 weeks at 0.09 mM extracellular Ca2+, as described in Materials and Methods. The immunocytochemical analysis of parallel cultures revealed that such monolayers contain approximately 65% of mature KC that are positive for the markers of terminal differentiation filaggrin, loricrin, involucrin, transglutaminase and cytokeratins 1/10/11, and approximately 35% of immature KC that are positive for the markers of proliferating cells Ki 67 and cytokeratins 5/6. The plate was installed on the preheated (37°C) stage of an inverted microscope and used in experiment. Note: the cells indicated with an asterisk were randomly selected to determine if the discharge of cytoplasmic material leads to cell shrinkage. (B) Same microscopic field 5 minutes after the culture received prewarmed (37°C) medium supplemented to contain 1 mM bromoacetylcholine and 0.1 mM atropine. Note: extensive cytoplasmic budding. (C) Same field 10 minutes after exposure. Note: the buds have grown in size and number, and many spheres have pinched off and float free. (D) Same field after the pinched off spheres were removed by collecting the chasing medium. Note: the cells indicated with an asterisk in A have condensed and shrunk to 91.2±4%, compared to their original size, taken as 100% (n=10). Bar, 50 μm.

Removal of drug-containing medium before the blebs started to detach aborted secretion, and the cytoplasmic blebbing reversed spontaneously. During the reversal, the cytoplasmic protrusions underwent retrocession, their contents was internalized by the cells within 20 minutes, and KC remained alive, as judged from results of the trypan blue dye exclusion test (data not shown).

Fate of the cells that have extruded their cytoplasmic material

To establish a functional link between the bleb extrusion phenomenon and its pharmacologic control on the one hand, and the cell regulation on the other, we followed the fate of the cells that had completed secretion. The discharge of cytoplasmic buds caused condensation and shrinkage of KC by approximately 10-15%, varying between different experiments. For instance, the size of the 10 intact KC that are marked with asterisks in Fig. 3A, taken as 100%, decreased to 91.2±4% after discharge (Fig. 3D). The discharge, however, did not alter the original polygonal shape of KC nor did it produce any visible alterations in the appearances of the nucleus and the nucleolus (Fig. 3D). However, the discharge resulted in the death of KC, as could be judged from their 100% staining with the trypan blue dye solution immediately after the discharge.

To determine if the cell death resulted from apoptosis, we analyzed DNA using the standard DNA fragmentation assay. The results demonstrated that the discharge of cytoplasmic buds is associated with the occurrence of the apoptosis pattern of DNA fragmentation in KC. Agarose gel electrophoresis of cellular DNA extracted from the KC that had extruded their cytoplasmic material showed a typical fragmentation ladder that was not seen in the DNA sample extracted from untreated cells (Fig. 4A).

Fig. 4.

Apoptotic pattern of DNA fragmentation in cultured human KC following secretion and characterization of the secreted material. The peripheral cytoplasmic blebbing and subsequent extrusion of the cytoplasmic buds was induced by treating a monolayer of human foreskin KC with 1 mM bromoacetylcholine and 0.1 mM atropine, exactly as described in the legend of Fig. 3. The extruded cytoplasmic buds were collected at different time points after extrusion by careful pipetting culture medium, controlled by observation through a phase-contrast microscope, and subjected to scanning electron microscopy or western blot analysis. The remaining monolayers were washed with drug-free culture medium containing 0.09 mM Ca2+, incubated in this medium for 24 hours and then used in the DNA fragmentation assay, as detailed in Materials and Methods. (A) Agarose gel electrophoresis of DNA extracted from the KC that have extruded cytoplasmic buds (experiment; lane 3) vs that of DNA extracted from the intact cells from the same cell donor that were grown under identical conditions and to the same degree of confluence as experimental cells (control; lane 2). The discharge of cytoplasmic material by experimental cells upon the secretagogue action of cholinergic drugs is associated with the appearance of DNA fragmentation ladder characteristic of apoptosis. The DNA markers are shown in lane 1. (B) Scanning electron microscopy revealed that at first the extruded spheres are simple, as illustrated by the sphere collected immediately after extrusion. Arrow indicates fibrilar material attached to the surface of the sphere, which may represent a remnant of the pedicle connecting this freshly separated sphere to the cell of its origin. Bar, 1.5 μm. (C) Later on, the surface of the extruded spheres becomes dimpled and infolded, as illustrated by this sphere which was collected approximately 5 minutes after extrusion. The spheres collected later than 10 minutes after extrusion could not be preserved for scanning electron microscopy examination because they broke and disintegrated (not shown). Bar, 1.5 μm. (D) Immunolocalization of filaggrin in the keratinocyte secretory product. Western blot of 15% SDS-PAGE-resolved proteins from normal human KC (positive control; lane 1), chasing medium containing cytoplasmic buds discharged by stimulated KC (experiment; lane 2) and chasing medium from non-treated keratinocyte culture (negative control; lane 3). The molecular masses in kDa are shown on the right. Arrow indicates a 37 kDa filaggrin band.

Fig. 4.

Apoptotic pattern of DNA fragmentation in cultured human KC following secretion and characterization of the secreted material. The peripheral cytoplasmic blebbing and subsequent extrusion of the cytoplasmic buds was induced by treating a monolayer of human foreskin KC with 1 mM bromoacetylcholine and 0.1 mM atropine, exactly as described in the legend of Fig. 3. The extruded cytoplasmic buds were collected at different time points after extrusion by careful pipetting culture medium, controlled by observation through a phase-contrast microscope, and subjected to scanning electron microscopy or western blot analysis. The remaining monolayers were washed with drug-free culture medium containing 0.09 mM Ca2+, incubated in this medium for 24 hours and then used in the DNA fragmentation assay, as detailed in Materials and Methods. (A) Agarose gel electrophoresis of DNA extracted from the KC that have extruded cytoplasmic buds (experiment; lane 3) vs that of DNA extracted from the intact cells from the same cell donor that were grown under identical conditions and to the same degree of confluence as experimental cells (control; lane 2). The discharge of cytoplasmic material by experimental cells upon the secretagogue action of cholinergic drugs is associated with the appearance of DNA fragmentation ladder characteristic of apoptosis. The DNA markers are shown in lane 1. (B) Scanning electron microscopy revealed that at first the extruded spheres are simple, as illustrated by the sphere collected immediately after extrusion. Arrow indicates fibrilar material attached to the surface of the sphere, which may represent a remnant of the pedicle connecting this freshly separated sphere to the cell of its origin. Bar, 1.5 μm. (C) Later on, the surface of the extruded spheres becomes dimpled and infolded, as illustrated by this sphere which was collected approximately 5 minutes after extrusion. The spheres collected later than 10 minutes after extrusion could not be preserved for scanning electron microscopy examination because they broke and disintegrated (not shown). Bar, 1.5 μm. (D) Immunolocalization of filaggrin in the keratinocyte secretory product. Western blot of 15% SDS-PAGE-resolved proteins from normal human KC (positive control; lane 1), chasing medium containing cytoplasmic buds discharged by stimulated KC (experiment; lane 2) and chasing medium from non-treated keratinocyte culture (negative control; lane 3). The molecular masses in kDa are shown on the right. Arrow indicates a 37 kDa filaggrin band.

Characterization of the substance secreted by KC

The secretory substance discharged by KC was examined by scanning electron microscopy to characterize its three-dimensional structure. The extruded spheres had smooth surface and were partially covered with some fibrilar material (Fig. 4B) which could represent a remnant of the pedicle connecting nascent spheres with the cell. The surfaces of the spheres harvested approximately 10-15 minutes after their detachment from the cells often looked dimpled and infolded (Fig. 4C), suggesting that the filaggrin-positive material had leaked out.

To determine whether the cytoplasmic buds extruded by stimulated KC contain keratohyalin material, we looked for filaggrin immunoreactivity. We (Grando et al., 1996a) and others (Lee et al., 1998; Li et al., 1995) have previously reported that confluent cultures of KC synthesize filaggrin in the amounts that are measurable by both immunohistochemistry and immunoblotting. Although filaggrin immunoreactivity could be observed in our keratinocyte monolayers (data not shown), rapid bursts of discharged spheres and high water solubility of filaggrin precluded its direct visualization in the extruded cytoplasmic buds using immunohistochemistry. Therefore, to test a hypothesis that KC extrude filaggrin, we collected chasing medium containing freshly extruded spheres, lyophilized samples, resolved proteins by SDS-PAGE, and determined filaggrin immunoreactivity by immunoblotting. The immunoreactivity of the proteins found in the chasing medium containing extruded spheres was compared with that of proteins of the chasing medium from non-treated monolayers (negative control) and of total keratinocyte extract (positive control). As seen in Fig. 4D, the cytoplasmic buds extruded by stimulated KC contained filaggrin. In both positive control and experimental samples, lanes 1 and 2, respectively, filaggrin band appeared at approximately 37 kDa, which agrees with the molecular mass of this protein reported previously (Fleckman et al., 1985; Grando et al., 1996a). The staining was abolished when the primary antibody was omitted (not shown). No protein bands were observed in the SDS-PAGE-resolved samples of lyophilized chasing medium from control, non-treated cells (Fig. 4D; lane 3). Investigation of other constituents of the excreted material was beyond the scope of this study.

Upward concentration gradient of free ACh in human epidermis

Having established that KC extrude keratohyalin in response to activation of their ACh receptors, we sought to identify the physiological mechanism that could explain how the secretagogue action of epidermal ACh commences precisely at the final stage of keratinocyte development in the epidermis. Previous immunohistochemical studies demonstrated the immunoreactivities of both cholinergic enzymes, ChAT and AChE, in human epidermis (Grando et al., 1993b), and measurements by an HPLC-method detected in the skin approximately 1000 pmol/g of free ACh (Klapproth et al., 1997). Since the presence of cholinergic enzymes in human epidermis has never been confirmed on the molecular level, we first performed RT-PCR experiments using RNA isolated from pure cultures of human KC and primers specific for ChAT and AChE. These PCR yielded products with the expected size, 291 bp for ChAT and 312 bp for AChE (Fig. 5A). No PCR products were amplified in either of two negative control experiments (Fig. 5A). Thus, the results of PCR experiments confirmed that human KC contain mRNA of ChAT, and AChE.

Fig. 5.

Demonstration of upward concentration gradient of free acetylcholine in human epidermis. (A) Agarose gel electrophoresis of RT-PCR products amplified from human keratinocyte mRNA using specific primers to ChAT and AChE (see Materials and Methods). The purified PCR products matched the sequences in GenBank. Lane 1 represents the experimental sample, lane 2 is a negative control for any residual genomic contamination, and lane 3 is another negative control sample in which the RNA template was omitted. (B through D) Immunohistochemical visualization of ChAT, AChE and ACh. The cryostats sections of normal human skin were fixed and immunostained as detailed in Materials and Methods. (B) Anti-ChAT antibody specifically stains KC throughout the epidermis. The stained epidermal compartments are: (i) the lowermost basal cell layer, comprised of a single row of polygonal epithelial cells residing on the epidermal basal membrane; (ii) the intermediate prickle cell layer, comprised of multiple rows of suprabasilar cells; and (iii) the uppermost granular cell layer, represented by 3 rows of large, flat cells which are covered by several layers of dead corneocytes, or flakes, comprising the stratum corneum (not stained). (C) Anti-AChE antibody immunostains predominantly the basal cell layer. (D) The distribution of immunostaining produced by anti-ACh antibody is reciprocal to that produced by anti-AChE antibody. The basal membrane is highlighted by a dotted line. Bar, 120 μm. (E) Relative amounts of ChAT, AChE, and ACh in different layers of human epidermis determined using a semi-quantitative immunofluorescence assay, as detailed in Materials and Methods.

Fig. 5.

Demonstration of upward concentration gradient of free acetylcholine in human epidermis. (A) Agarose gel electrophoresis of RT-PCR products amplified from human keratinocyte mRNA using specific primers to ChAT and AChE (see Materials and Methods). The purified PCR products matched the sequences in GenBank. Lane 1 represents the experimental sample, lane 2 is a negative control for any residual genomic contamination, and lane 3 is another negative control sample in which the RNA template was omitted. (B through D) Immunohistochemical visualization of ChAT, AChE and ACh. The cryostats sections of normal human skin were fixed and immunostained as detailed in Materials and Methods. (B) Anti-ChAT antibody specifically stains KC throughout the epidermis. The stained epidermal compartments are: (i) the lowermost basal cell layer, comprised of a single row of polygonal epithelial cells residing on the epidermal basal membrane; (ii) the intermediate prickle cell layer, comprised of multiple rows of suprabasilar cells; and (iii) the uppermost granular cell layer, represented by 3 rows of large, flat cells which are covered by several layers of dead corneocytes, or flakes, comprising the stratum corneum (not stained). (C) Anti-AChE antibody immunostains predominantly the basal cell layer. (D) The distribution of immunostaining produced by anti-ACh antibody is reciprocal to that produced by anti-AChE antibody. The basal membrane is highlighted by a dotted line. Bar, 120 μm. (E) Relative amounts of ChAT, AChE, and ACh in different layers of human epidermis determined using a semi-quantitative immunofluorescence assay, as detailed in Materials and Methods.

The concentration of free ACh is a function of its synthesis by ChAT and hydrolysis by AChE. By indirect immunofluorescence, ChAT was found distributed equally throughout the entire epidermis (Fig. 5B). The degrading enzyme AChE was seen predominantly in the basal cell layer (Fig. 5C). This distribution of ChAT and AChE suggested that free ACh is accumulated in the uppermost epidermal compartment. Staining of the epidermis with an antibody specific to ACh confirmed the above assumption. ACh staining was abundant in the granular layer, whereas the lowermost basal layer remained unstained (Fig. 5D). Omitting primary antibody or replacing it with an isotype-control antibody abolished the fluorescent staining for each cholinergic molecule (data not shown).

The relative amounts of ChAT and AChE and free ACh in three layers of viable KC, the basal, the spinous, and the granular, were also characterized using semi-quantitative immunofluorescence assay (Fig. 5E). The relative amounts of ChAT in different epidermal layers did not significantly (P>0.05) differ from each other. In marked contrast, the relative amount of AChE decreased from 131.8±5, characteristic of basal KC, to 40.5±5 found in granular KC (P<0.05). Accordingly, the relative amount of free ACh increased from 9.8±4 to 123.7±5 (P<0.05), peaking at the granular layer (Fig. 5E).

The repertoire of cholinergic receptors expressed by terminally differentiated KC

To better understand the biochemical mechanism leading to extrusion of the cytoplasmic buds, we sought to identify the types of cholinergic receptors which can mediate the secretagogue action of ACh in the upper epidermal compartment. The pharmacological profile of the secretion regulation suggested a role for activation of the nicotinic pathway and inactivation of muscarinic receptors. To characterize in detail the receptor-mediated pathways of ACh signaling during the granular cell-corneocyte transition, we used a semi-quantitative immunofluorescence assay with anti-receptor antibodies. Based on the results of our previous studies showing the presence of α7 and α9 nicotinic and m1 and m4 muscarinic receptors in the superficial layers of the stratified squamous epithelium (Ndoye et al., 1998; Nguyen et al., 2000b; Zia et al., 2000), we used antibodies to the above four receptor types to characterize the cholinergic receptor repertoire of the innermost, middle, and uppermost KC in the granular layer.

In human epidermis, the bulk of α7 immunoreactivity was localized to the cell membranes of KC comprising the granular layer (Fig. 6A). The intensity of specific staining of KC with α9 antibody increased from the lowermost to the uppermost epidermal compartments (Fig. 6B). Semi-quantitative immunofluorescence assay revealed that the relative amount of α7 increases significantly (P<0.05) and that of α9 insignificantly (P>0.05) with each step of terminal differentiation of KC in the epidermis, i.e. during transition of a keratinocyte from the innermost to the uppermost row of the stratum granulosum (Fig. 6C).

Fig. 6.

Characterization of nicotinic ACh receptors expressed by terminally differentiated human KC. (A and B) Immunohistochemical visualization of α7 (A) and α9 (B) ACh receptors in the epidermis. Both anti-receptor antibodies produce a fishnet-like staining of the epidermis, indicating the presence of the receptors on the cell membranes. The α7 antibody stains mainly the cell membrane of the granular cells. Although immunostaining for α9 is seen throughout the epidermis, it is most abundant in the superficial epidermal compartment, with the uppermost 2-3 rows of the prickle cell layer and the entire granular cell layer stained most brightly. The immunostaining was eliminated when the primary rabbit anti-receptor antibodies were omitted and when the rabbit antiserum specific for each ACh receptor type was preincubated with the peptides used for immunization. No immunostaining was observed when the KC were treated with pre-immune sera obtained from the same rabbits (not shown). Bars, 25 μm. (C) Semi-quantitative analysis of relative amounts of keratinocyte α7, and α9 ACh receptors visualized in different rows of the stratum granulosum of human epidermis by the specific antibodies. (D through F) Cryostat sections of freshly frozen colonies of first passage human foreskin KC grown for two days on coverslips were fixed, and immunostained with rabbit anti-α9 antibody. (D) The α9-positive KC in the culture grown at low, 0.09 mM, extracellular Ca2+. Bar, 25 μm. (E) An increased number of α9-positive KC in the culture incubated for 24 hours in the medium containing high, 1.8 mM, Ca2+. Bar, 25 μm. (F) A high-power view of the stained monolayer that had been incubated at high, 1.8 mM, extracellular Ca2+. Note: accumulation of anti-α9 antibody on the cell membrane areas associated with the sites of cell-to-cell contacts. Bar, 15 μm.

Fig. 6.

Characterization of nicotinic ACh receptors expressed by terminally differentiated human KC. (A and B) Immunohistochemical visualization of α7 (A) and α9 (B) ACh receptors in the epidermis. Both anti-receptor antibodies produce a fishnet-like staining of the epidermis, indicating the presence of the receptors on the cell membranes. The α7 antibody stains mainly the cell membrane of the granular cells. Although immunostaining for α9 is seen throughout the epidermis, it is most abundant in the superficial epidermal compartment, with the uppermost 2-3 rows of the prickle cell layer and the entire granular cell layer stained most brightly. The immunostaining was eliminated when the primary rabbit anti-receptor antibodies were omitted and when the rabbit antiserum specific for each ACh receptor type was preincubated with the peptides used for immunization. No immunostaining was observed when the KC were treated with pre-immune sera obtained from the same rabbits (not shown). Bars, 25 μm. (C) Semi-quantitative analysis of relative amounts of keratinocyte α7, and α9 ACh receptors visualized in different rows of the stratum granulosum of human epidermis by the specific antibodies. (D through F) Cryostat sections of freshly frozen colonies of first passage human foreskin KC grown for two days on coverslips were fixed, and immunostained with rabbit anti-α9 antibody. (D) The α9-positive KC in the culture grown at low, 0.09 mM, extracellular Ca2+. Bar, 25 μm. (E) An increased number of α9-positive KC in the culture incubated for 24 hours in the medium containing high, 1.8 mM, Ca2+. Bar, 25 μm. (F) A high-power view of the stained monolayer that had been incubated at high, 1.8 mM, extracellular Ca2+. Note: accumulation of anti-α9 antibody on the cell membrane areas associated with the sites of cell-to-cell contacts. Bar, 15 μm.

In keratinocyte cultures, the abundant expression of α7 by differentiated cells was reported by us previously (Zia et al., 2000). In this study, we used α9 antibody to determine the effect of differentiation on the expression of α9 by cultured KC. As expected, rapid differentiation due to incubation of cells in the medium containing high, 1.8 mM, Ca2+ which launches keratinocyte differentiation program (Sharpe et al., 1993), dramatically increased the number of KC stained specifically with α9 antibody (Fig. 6D,E). The receptors were seen accumulated at the areas of the cell membrane that mediated cell-to-cell contacts (Fig. 6F). Thus, KC comprising confluent monolayers which were used to elicit filaggrin secretion express α7, and α9 nicotinic receptors.

In parallel series of immunofluorescence experiments, we investigated the effect of differentiation on the expression of m1 and m4 muscarinic receptors. The graded staining pattern of m1 expression in epidermis was found to be very similar to that of α7 and the results of semi-quantitative analysis of fluorescence staining demonstrated that m1 is abundant in the outermost granular KC (Fig. 7A,C). In marked contrast, the immunoreactivity of m4 decreased significantly (P<0.05) as KC advanced from the innermost to the uppermost row of the stratum granulosum (Fig. 7B,C). Both m1 and m4 were also demonstrated on the cell-surfaces of cultured KC, predominantly at the sites of cell-to-cell contacts (Fig. 7D,E). As seen in Fig. 7D, in the central part of the colony (upper right corner), which is composed of differentiated KC forming stable intracellular junctions (Krawczyk and Wilgram, 1973), the immunostaining for the m4 receptor was much less abundant compared to the more peripheral parts of the colony.

Fig. 7.

The muscarinic receptors expressed by terminally differentiated KC and the changes in [Ca2+]i mediated by these receptors. (A through E) Immunolocalization of keratinocyte muscarinic ACh receptors. Rabbit polyclonal antibodies raised to m1 (A,E) and m4 (B,D), previously found to be expressed by human KC, were used to immunolocalize these receptors. Cryostat sections of freshly frozen specimens of normal human skin (A,B) and colonies of first passage human foreskin KC grown on coverslips (D,E) were fixed, and immunostained with the receptor specific antibodies, as detailed in Materials and Methods. (A) An upward pattern of m1 expression in human epidermis. Anti-m1 antibody produces a fishnet-like, cell membrane staining of KC in the epidermis, which is most abundant in the superficial rows of the granular cell layer. Bar, 25 μm. (B) Graded pattern of m4 expression in human epidermis. The bulk of m4 immunoreactivity is localized to the lowermost rows of the prickle cell layer, whereas the granular cell layer stains weakly. Bar, 25 μm. (C) Relative amounts of m1 and m4 expressed by KC comprising the three different rows of the granular cell layer. (D) Visualization of m4 in cultured KC grown at 0.09 mM Ca2+ in the growth medium. The m4 immunoreactivity gradually decreases toward the central area of the colony comprised by more differentiated KC, situated at the upper right corner of the photograph. Bar, 50 μm. (E) Clusters of m1 receptors on the plasma membrane of cultured KC incubated at high, 1.8 mM, concentration of Ca2+ in the growth medium. Anti-m1 antibody visualizes the m1 muscarinic ACh receptor subtype as distinct bright dots decorating the perinuclear and more peripheral areas the of the cell membrane of two KC closely opposed to one another. Bar, 15 μm. (F through H) Fluctuations of [Ca2+]i levels in first passage intact human foreskin KC exposed to muscarinic antagonists tropicamide, pirenzepine, and atropine. First passage, intact human foreskin KC were subcultured into 2-well glass chamber slides, fed with culture medium containing 0.09 mM Ca2+, incubated in a humid, 5% CO2 incubator at 37°C to allow the cells to adhere to the dish bottom and spread their cytoplasm. The cells were then washed and the [Ca2+]i was measured in twenty individual cells, as described in Materials and Methods. The data are an average of the [Ca2+]icontinuously measured in twenty KC at the magnification ×200. The results were reproduced in at least 3 independent experiments. (F) The selective m4 receptor antagonist tropicamide (0.3 mM) elicits a transient increase in [Ca2+]i. (G) Pretreatment of KC with the m1 receptor antagonist pirenzepine (0.3 mM) abolishes the tropicamide (0.3 mM)-induced increase in [Ca2+]i. (H) The effect of atropine (0.5 mM) is similar to that of tropicamide.

Fig. 7.

The muscarinic receptors expressed by terminally differentiated KC and the changes in [Ca2+]i mediated by these receptors. (A through E) Immunolocalization of keratinocyte muscarinic ACh receptors. Rabbit polyclonal antibodies raised to m1 (A,E) and m4 (B,D), previously found to be expressed by human KC, were used to immunolocalize these receptors. Cryostat sections of freshly frozen specimens of normal human skin (A,B) and colonies of first passage human foreskin KC grown on coverslips (D,E) were fixed, and immunostained with the receptor specific antibodies, as detailed in Materials and Methods. (A) An upward pattern of m1 expression in human epidermis. Anti-m1 antibody produces a fishnet-like, cell membrane staining of KC in the epidermis, which is most abundant in the superficial rows of the granular cell layer. Bar, 25 μm. (B) Graded pattern of m4 expression in human epidermis. The bulk of m4 immunoreactivity is localized to the lowermost rows of the prickle cell layer, whereas the granular cell layer stains weakly. Bar, 25 μm. (C) Relative amounts of m1 and m4 expressed by KC comprising the three different rows of the granular cell layer. (D) Visualization of m4 in cultured KC grown at 0.09 mM Ca2+ in the growth medium. The m4 immunoreactivity gradually decreases toward the central area of the colony comprised by more differentiated KC, situated at the upper right corner of the photograph. Bar, 50 μm. (E) Clusters of m1 receptors on the plasma membrane of cultured KC incubated at high, 1.8 mM, concentration of Ca2+ in the growth medium. Anti-m1 antibody visualizes the m1 muscarinic ACh receptor subtype as distinct bright dots decorating the perinuclear and more peripheral areas the of the cell membrane of two KC closely opposed to one another. Bar, 15 μm. (F through H) Fluctuations of [Ca2+]i levels in first passage intact human foreskin KC exposed to muscarinic antagonists tropicamide, pirenzepine, and atropine. First passage, intact human foreskin KC were subcultured into 2-well glass chamber slides, fed with culture medium containing 0.09 mM Ca2+, incubated in a humid, 5% CO2 incubator at 37°C to allow the cells to adhere to the dish bottom and spread their cytoplasm. The cells were then washed and the [Ca2+]i was measured in twenty individual cells, as described in Materials and Methods. The data are an average of the [Ca2+]icontinuously measured in twenty KC at the magnification ×200. The results were reproduced in at least 3 independent experiments. (F) The selective m4 receptor antagonist tropicamide (0.3 mM) elicits a transient increase in [Ca2+]i. (G) Pretreatment of KC with the m1 receptor antagonist pirenzepine (0.3 mM) abolishes the tropicamide (0.3 mM)-induced increase in [Ca2+]i. (H) The effect of atropine (0.5 mM) is similar to that of tropicamide.

These results indicate that during the latest stages of their differentiated the m4 muscarinic receptor-coupled signaling pathway becomes inactive and in their responses to ACh, KC use the m1-coupled intracellular signaling, in addition to the α7-, and α9 nicotinic receptor-mediated pathways.

The muscarinic receptor-coupled pathway of ACh signaling in terminally differentiated KC increases [Ca2+]i

We further investigated the receptor-mediated pathway of ACh-induced keratohyaline secretion by measuring cholinergic effects on Ca2+ metabolism, because ACh uses Ca2+ as a second messenger of its biological effects on KC (reviewed by Grando, 1997). We have previously demonstrated that stimulation of keratinocyte nicotinic receptors elicits both transmembrane influx and intracellular raise of Ca2+ (Grando et al., 1996a; Zia et al., 2000), both of which can be abolished by mecamylamine. In this study, we examined the fluctuations in [Ca2+]i levels in KC exposed to the muscarinic antagonists capable of facilitating extrusion of the cytoplasmic buds. We employed the ratiometric analysis of cells loaded with Fura-2 to quantitate the total concentration of free cytoplasmic Ca2+ which can originate from the mobilization of intracellular stores and from the influx through Ca2+-channels in the plasma membrane (Boess et al., 1990; Grynkiewicz et al., 1985). Tropicamide, the specific m4 receptor antagonist (Lazareno and Birdsall, 1993), rapidly increased the concentration of [Ca2+]i above the basal level (Fig. 7F). The [Ca2+]i peaked at more than 4 times its baseline concentration, but then rapidly dropped to a concentration below the initial basal level and established a new baseline. Since blocking the intracellular signaling pathway from the m4 receptor elicited an increase in the [Ca2+]i, constant activation of this ‘inhibitory’ receptor subtype with endogenous ACh might be opposing a ‘stimulatory’ pathway that constantly increases the [Ca2+]i through an odd-numbered receptor. Therefore, we hypothesized that blocking a ‘stimulatory,’ odd-numbered receptor subtype should prevent the tropicamide-induced rise in [Ca2+]i. Since KC secreting keratohyaline material in the epidermis express m1 receptor, we exposed cultured cells to the m1 inhibitor pirenzepine (Kebabian and Neumeyer, 1994). As expected, pirenzepine prevented the tropicamide-induced increase in [Ca2+]i (Fig. 7G), indicating that m1 mediates the ‘stimulatory’ pathway of ACh signaling that increases [Ca2+]i in KC. The effect of the tropicamide congener atropine on [Ca2+]i was very similar to that of tropicamide (Fig. 7H). The concentration of [Ca2+]i in KC exposed to atropine increased rapidly from the baseline of 21 nM to 127 nM followed by a rapid decline to the initial level.

These results indicate that rise of [Ca2+]i is an essential element in the biochemical mechanism mediating secretion of keratohyaline material by KC.

In this study, we demonstrate for the first time that when terminally differentiated KC undergo transition into devitalized corneocytes they spontaneously extrude bits of their cytoplasm containing keratohyaline material, featured by filaggrin, as well as various randomly congregated organelles. Since all terminally differentiated KC in culture show peripheral membrane blebbing with subsequent extrusion of cytoplasmic buds, the phenomenon described in this study may, in effect, represent a novel and essential step of the programmed cell death in the epidermis. The secretagogue for this novel keratinocyte secretion is autocrine and paracrine ACh activating the m1 muscarinic and the α7, and the α9 nicotinic receptors that modulate [Ca2+]i levels in KC.

The purpose of this study was to investigate how terminally differentiated KC rid of most of their cell volume during programmed cell death, or cornification. Cornification is one of the most intriguing aspects of keratinocyte biology. It is a final step of programmed cell death in the epidermis, which, therefore, is considered by many (Dale et al., 1997; Haake and Polakowska, 1993; McCall and Cohen, 1991), but not all (Gandarillas et al., 1999), to be a form of apoptosis. Since apoptosis is often associated with peripheral membrane blebbing (reviewed by Majno and Joris, 1995), we looked specifically at the morphology of terminally differentiated KC in vivo and in vitro. We were encouraged by the fact that one classic ultrastructural study clearly showed an outermost granular cell projecting a large cytoplasmic bud filled with keratohyaline masses (see Fig. 14 by Odland, 1971). In a series of electron microscopic and phase-contrast experiments we demonstrated that the granular cell-corneocyte transition is indeed associated with extrusion of cytoplasmic buds containing filaggrin as well as some cellular organelles. Although filaggrin-positive keratohyaline material was present in all buds, the process of keratinocyte secretion described in this study is apparently not specific for only filaggrin secretion. The spheres also contained lysosomes, mitochondria, glycogen granules, parts of rough or smooth endoplasmic reticulum and lipids. Therefore, it appeared that this novel mechanism for volume loss, which provides for secretion of keratohyaline material, together with the release of lamellar bodies (Matoltsy, 1966), play an important role in the process of cornification.

In the past, keratohyaline secretion by KC might be unnoticed for the following reasons. First, the transition of granular KC into corneocytes is abrupt (Lavker and Matoltsy, 1970), and cytoplasmic budding of rapidly dying KC is a sporadic finding on electron microscopy which can be easily overlooked. Second, cytoplasmic buds can not be clearly distinguished in vertical sections of epidermis because the buds are located in the same plane as the granular KC that extrude them. Third, the cytoplasmic budding in vitro occurs in organotypic crowns comprised of terminally differentiated KC, which are located in the microscope focus plane that is above the cell monolayer, a subject of the microscopic examination. Fourth, filaggrin is not detectable in the intercellular spaces of epidermis upon antibody binding because during the staining procedure this highly soluble protein is washed away. Since the immunoreactivity of non-secreted filaggrin could be easily demonstrated in corneocytes, a possibility that only a fraction rather than the total amount of filaggrin becomes associated with fibrilar proteins was ignored. In fact, filaggrin extruded by granular KC may leak downward between spinous KC because electron microscopic stains for glycosylated moieties, demonstrate electron dense material within intracellular spaces (Wolff and Schreiner, 1968). We therefore propose that products of enzymatic degradation of keratohyaline material contribute to formation of the glycocalyx, the ‘intercellular cement substance’ of epidermis.

The presence of natural humectants, such as filaggrin degradation products (Scott and Harding, 1981; Scott and Harding, 1986), throughout epidermis would explain its exceptional plasticity and resilience. Furthermore, this would explain how the epidermis manages to avoid osmotic shock which would have occurred if all hydroscopic humectants of the skin were present exclusively in the stratum corneum, as it is currently held with respect to the natural moisturizing factor (NMF) (Rawlings et al., 1994). Since the permeability barrier formed by lamellar body lipids underneath the stratum corneum is actually permeable to water to the extent that allows substantial transepidermal traffic of hydrophilic drugs (Williams et al., 1992), the resulting osmotic pressure would then absorb out all intercellular fluid from the underlying living epidermis, which is clearly not the case. The stratum corneum does soak water, but from the outside rather than from the inside, which explains why the surface of the skin swells and whitens only upon prolonged bathing. We therefore propose that there exist two types of NMFs: the ‘external’ NMF, which is present in the stratum corneum, and the ‘internal’ NMF, which fills intracellular spaces of living epidermis. The external NMF is associated with the corneocytes because it is used for aggregating keratin filaments at the final stages of keratinocyte differentiation (Brody, 1960). This NMF absorbs water from the environment (Scott and Harding, 1986). The osmotic pressure generated by the external NMF is balanced by the internal NMF which derives from the fraction of the filaggrin that is secreted by terminally differentiated KC into the intracellular spaces of living epidermis. The internal NMF may absorb water from the underlying dermis and keep it within the epidermis, thus providing a mechanism for maintenance of epidermal plasticity in a hot and dry environment. The water absorbed by internal NMS may also play a role in maintaining soft texture of human skin, thus contributing to its ‘young-looking’ appearance. Since exposure to pure nicotine results in rapid increase of filaggrin content in human KC (Grando et al., 1996a), the amount of internal NMF may be regulated, at least in part, via the nicotinic limb of the ACh axis of keratinocyte differentiation. Overstimulation of nicotinic receptor-mediated pathways is known to produce antagonist-like biological effects, due to receptor desensitization (Yang and Buccafusco, 1994; Zia et al., 1997). Therefore, exhaustion of the nicotinic receptor-mediated regulatory pathway of internal NMS production may offer a novel explanation of early appearance of premature aged, i.e., thin, dry, pale, rough and wrinkled, or simply ‘cigarette,’ skin in nicotine users (Smith and Fenske, 1996).

The ability of nicotine to increase filaggrin content in KC may result from activation of the pathway(s) coupled to acceleration of the genetically-determined program of keratinocyte development in the epidermis. Indeed, activation of the α7 signaling pathways has been proposed to mediate ACh-induced differentiation of human KC (Grando et al., 1996a; Zia et al., 2000). On the other hand, KC treated with a muscarinic antagonist alone exhibited only a reversible cytoplasmic blebbing but did not extrude their cytoplasmic buds, indicating that a combination of a cholinergic agonist and a muscarinic antagonist was required to trigger expulsion of the cytoplasmic buds. Thus, to trigger keratinocyte secretion, the nicotinic pathways of ACh signaling should have been activated at the background of inhibition of muscarinic pathway(s).

The inability of the endogenously produced ACh to trigger secretion via uninhibited nicotinic pathways could be due to rapid degradation of ACh by AChE, which is expressed abundantly on the cell surfaces of cultured human KC (Grando et al., 1993b). Therefore, we exposed KC to AChE-resistant ACh analogs, such as nicotine, cytisine, carbachol and bromoacetylcholine (Brailoiu and Van Der Kloot, 1996; DiPalma, 1994). Only carbachol and bromoacetylcholine could elicit the discharge of cytoplasmic buds. Inability of nicotine or cytisine to trigger the discharge was reminiscent of a failure of this classic nicotinic agonists to elicit current responses in Xenopus laevis oocytes with functionally expressed α9 (Elgoyhen et al., 1994). Although both of these classic nicotinic agonists can compete with endogenous ACh for binding to α9-containing receptors, they exhibit an antagonist rather than agonist effect on this novel type of cholinergic receptors (Verbitsky et al., 2000) which was first cloned by us from human tissues using oral keratinocyte mRNA (Nguyen et al., 2000b). Furthermore, the detachment of secretory buds could be blocked in the presence of strychnine, which blocks α9 receptors (Elgoyhen et al., 1994), suggesting that activation of the α9 pathway of ACh signaling is required to accomplish the excretory phase of keratinocyte secretion. Indeed, activation/blockade of α9 receptors has been recently demonstrated to produce reciprocal changes in shape and motility of human oral KC in vitro (Nguyen et al., 2000b). Taken together, these observations suggest that activation of the α7-coupled pathway provides for filaggrin production and accumulation in the cytoplasm of terminally differentiated KC in the form of keratohyaline granules, whereas activation of the α9-coupled pathway facilitates extrusion of this material during the granular cell-corneocyte transition.

Extrusion of the cytoplasmic buds by KC looks like a form of exocrine secretion. The three-dimensional image of the secretory substance of KC (Fig. 4B,C) is very similar to that released from the secretory cells of an apocrine gland (Kurosumi and Kawabata, 1976). Apocrine mechanism of secretion, however, implies that following extrusion of bits of apical cytoplasm, the secretory cells are able to not only seal the plasmalemma but also to accumulate and extrude the secretory product anew (Geneser, 1986). Since we observed that KC die after extrusion pieces of their cytoplasm, this is not the apocrine type of secretion. Is keratohyaline secretion by KC of holocrine type then? The term ‘holocrine’ stems from the Greek holos, which means entire or complete, and the secretory mechanism involves cell rupture followed by discharge of accumulated product, such as lipid in the sebaceous glands of the skin (Thody and Shuster, 1989). This is also different from extrusion of bits of the cytoplasm by KC. Merocrine secretion is a third, and the last, among major mechanisms of secretion (Geneser, 1986). It describes process of exocytosis where the secretory product is discharged without a loss of cell substance. Therefore, extrusion of keratohyaline material does not meet any of the above three criteria for classical secretory mechanisms.

Spontaneous extrusion of organelle-rich endoplasm containing keratohyaline culminates programmed death of KC in the epidermis, suggesting that this secretion is a part of cell suicide, or apoptosis. This conjecture is supported by the results of DNA fragmentation assay showing typical ‘apoptotic’ ladder. Apoptosis can be associated with peripheral cytoplasmic budding, also termed zeiosis, which may provide a mechanism for a loss of cell volume by apoptotic cells (reviewed by Kerr et al., 1972; Majno and Joris, 1995). Extrusion of cytoplasmic buds observed in our in vitro studies also was followed by cell shrinkage by approximately 10-15%, indicating that this secretion mediates volume loss by KC. Furthermore, by analogy with expulsion of cytoplasmic protrusions by the cells undergoing zeiosis (Godman et al., 1975), the cytoplasmic knobs of KC also pinched off, and the three-dimensional images of individual zeiotic blebs and keratinocyte spheres look alike. Therefore, it is plausible to speculate that keratohyaline secretion is a form of apoptosis. In certain cells types, zeiotic blebbing is associated with, and therefore may specifically provide for, an accomplishment of the last and most important biological function by doomed cells, such as degranulation of polymorphonuclear leukocytes (Bignold and Ferrante, 1988). Therefore, we propose that extrusion of endoplasm by terminally differentiated KC represents a novel and distinctive form of cell suicide that defines zeiosis-like expulsion of the secretory products leading to dramatic loss of cell volume. We suggest to name this biological phenomenon the ‘apoptotic secretion.’

The apoptotic secretion of KC is controlled by the cytotransmitter ACh, and ACh may use Ca2+ as a second messenger of its secretagogue action on KC. Recently, a nonneuronal ACh has emerged as a local hormone operating in various types of the epithelium where it controls a plethora of cellular functions in an autocrine and paracrine way (reviewed by Grando, 1997; Wessler et al., 1998). For instance, cholinergic system of rat cholangiocytes modulates both apoptosis and secretion (LeSage et al., 1999). Stimulation of nicotinic receptors can trigger apoptosis of thymic epithelial cells (Rinner et al., 1999), whereas activation of the m3, but not m1, muscarinic receptor can block apoptosis of cerebellar granule neurons (Yan et al., 1995). This phenomenon may be explained based on the fact that each class of ACh receptors exhibits inverse effects on Ca2+ metabolism. The presence of Ca2+ was required to elicit cytoskeletal changes mediating the secretagogue action of the cholinergic agonist carbachol in parotid acinar cells (Perrin et al., 1992). An increase of cytosolic Ca2+ can be also associated with both zeiotic blebbing of apoptotic cells (Jewell et al., 1982) and extrusion of secretory products by secretory cells (Brown and Chew, 1989), both of which are mediated by high Ca2+-induced alterations in the cytoskeleton (Burgoyne et al., 1991; Orrenius and Nicotera, 1994). In KC, too, Ca2+ plays a critical role in modification of the cytoskeleton and secretion of lamellar bodies (Lewis et al., 1994; Menon et al., 1994). We have previously demonstrated that activation of the nicotinic ACh receptors expressed by the epithelial cells comprising the epidermis and lining the upper respiratory tract elicits both transmembrane flux of Ca2+ and rise in [Ca2+]i (Grando et al., 1996a; Zia et al., 2000; Zia et al., 1997).

The stimulatory effect of ACh on Ca2+ influx in the KC, mediated by its nicotinic action, is balanced by an inhibitory effect, mediated by its muscarinic action (Grando and Horton, 1997). Simultaneous activation of both pathways may produce a kind of a yin-yang regulatory balance, and a fine tuning of the biological response may be achieved through an interplay between the odd-and even-numbered muscarinic receptors. Activation of the ‘stimulatory’ m1, m3, and m5 receptors can increase [Ca2+]i by releasing Ca2+ from the endoplasmic reticulum, whereas activation of the ‘inhibitory’ m2 and m4 receptors can decrease [Ca2+]i by suppressing transmembrane Ca2+ flux (reviewed by Caulfield, 1993; Jones, 1993). The relative amount of m1 increases steadily during the transition of KC through the granular layer, reaching the peak at the site of apoptotic secretion, whereas the changes in the relative amounts of m4 are just as opposite, with almost complete disappearance of this receptor from the outermost granular KC. These shifts in the repertoire of keratinocyte muscarinic receptors can increase [Ca2+]i because both the transmembrane influx of Ca2+ and the release of sequestered Ca2+ from intracellular stores become unopposed. Since the m1 antagonist pirenzepine prevented tropicamide-induced increase in [Ca2+]i, endogenous ACh must release Ca2+ from intracellular stores via the m1-coupled pathway. Indeed, coupling of m1 to Ca2+ mobilization from intracellular stores has been demonstrated in a variety of cell types (Boddeke et al., 1992; Dolezal et al., 1997). In our study, atropine induced a rise of [Ca2+]i which was very similar to that induced by its congener tropicamide, indicating that keratinocyte m4 was the predominant target for atropine in its secretagogue-like effect on KC.

The repertoire of cholinergic enzymes and receptors expressed by terminally differentiated KC provides for an unopposed secretagogue action of epidermal free ACh at the stage of granular cell-corneocyte transition. A physiologic upward concentration gradient of Ca2+, which correlates with the level of keratinocyte differentiation and peaks at the stratum granulosum-stratum corneum interface, has been demonstrated in epidermis (Mauro et al., 1998; Menon et al., 1992). In this study, we observed that the concentration of free ACh is also higher in the upper compared to the lower epidermis, which confirms our previous speculation (Grando, 1997). At the final stage of their development in the epidermis, KC have already lost the degrading enzyme AChE that can oppose the secretagogue action of ACh, but express abundantly the ACh synthesizing enzyme ChAT. Thus, an excess of both free ACh and Ca2+ in the upper epidermal compartment may be needed to provide granular KC with respectively the first and second messengers required for triggering apoptotic secretion.

Spontaneous excretion of the cytoplasmic buds that was detected by us in both skin specimens and intact cell cultures commenced at the latest stage of keratinocyte development, i.e., in the uppermost row of the granular cell layer and in the organotypic structures crowning monolayers of aged cultures, respectively. The secreting KC were terminally differentiated, because they were positive for the markers of terminal differentiation filaggrin, loricrin, involucrin, transglutaminase and cytokeratins 1/10/11, and negative for the Ki 67 antigen and the cytokeratins 5/6. Treatment of monolayers comprised of both mature and immature KC with certain combinations of cholinergic drugs mimicked apoptotic secretion, indicating that even non terminally differentiated KC can extrude bids of their cytoplasm upon appropriate cholinergic stimulation. The drug-induced excretion of the cytoplasmic buds, however, is different from the naturally occurring apoptotic secretion because it is not related to the cell cycle. The induction of budding in drug-treated monolayers was possible because the ‘secretory machinery’ that mediates apoptotic secretion in terminally differentiated KC is functional, albeit inactive, at earlier stages of keratinocyte development. This secretory machinery is represented by ACh receptors and the biochemical pathways that couple these receptors to an increase of [Ca2+]i. The apoptotic secretion does not occur precociously at earlier stages of keratinocyte differentiation because ACh signaling through ‘pro-secretory’ receptors, such as m1 muscarinic and α7 and α9 nicotinic receptors, is counterbalanced due to simultaneous activation of the ‘anti-secretory’ receptors lowering [Ca2+]i, such as m4 muscarinic receptor. Disappearance of m4 from the cell membrane of KC during the granular cell-corneocyte transition allows an unopposed activation of m1, α7 and α9 on cell surfaces by ACh, which may disturb the physiologic equilibrium of oscillations of the [Ca2+]i. Notably, α7 and α9 nicotinic receptors can both mediate transmembrane flux of Ca2+ and facilitate raise of [Ca2+]i (Delbono et al., 1997; Jagger et al., 2000; Seguela et al., 1993; Wikstrom et al., 1998). The imbalance between the stimulatory and inhibitory effects of ACh can provide for a sudden rise in [Ca2+]i, launching extrusion of cytoplasmic buds. Therefore, since KC comprising confluent monolayers in vitro also express the pro-secretory receptors m1, α7, and α9 and since the anti-secretory m4 receptor can be selectively blocked by a pharmacologic antagonist, it is possible to induce extrusion of the cytoplasmic buds from non terminally differentiated KC. Thus, it is an intracellular biochemical event, triggered by activation of a certain combination of ACh receptors expressed by a keratinocyte at a particular stage of its development, that is responsible for launching of the apoptotic secretion.

We conclude that:

  1. he dramatic volume loss that is associated with devitalization of KC in epidermis at the final stage of their programmed cell death represents apoptotic secretion, a form of cell suicide mediated by ACh-induced sporadic discharge of randomly congregated cell components and keratohyaline material into intercellular space of epidermis. We propose that filaggrin-containing portion of the secretory product may become a part of glycocalyx and serve as an internal NMF that counterbalances the osmotic pressure imposed by the external NMF located in the stratum corneum;

  2. spontaneous secretion of cell components by terminally differentiated KC is accomplished due to an unopposed, simultaneous activation of the m1-, α7-, and α9-mediated pathways of the keratinocyte cholinergic system with an autocrine and paracrine ACh which uses Ca2+ as a second messenger of its secretagogue action; and further studies are needed to establish physiologic relevance and significance of this novel form of keratinocyte secretion.

This work was supported in parts by the NIH grant AR42955 and a research grant from the Unilever Research-USA to Dr Grando.

Bignold
,
L. P.
and
Ferrante
,
A.
(
1988
).
Effects of cytochalasin B, N-formyl peptide and plasma on polarisation, zeiosis (blebbing) and degranulation of polymorphonuclear leukocytes in suspension
.
Cell Biol Int. Rep
.
12
,
195
203
.
Boddeke
,
H. W.
,
Buttini
,
M.
,
Lichtsteiner
,
M.
and
Enz
,
A.
(
1992
).
M1 muscarinic receptors mediate intracellular calcium release in NB-OK1 human neuroblastoma cells
.
Arch. Pharmacol
.
346
,
255
261
.
Boess
,
F. G.
,
Balasubramanian
,
M. K.
,
Brammer
,
M. J.
and
Campbell
,
I. C.
(
1990
).
Stimulation of muscarinic acetylcholine receptors increases synaptosomal free calcium concentration by protein kinase-dependent opening of L-type calcium channels
.
J. Neurochem
.
55
,
230
236
.
Brailoiu
,
E.
and
Van Der Kloot
,
W.
(
1996
).
Bromoacetylcholine and acetylcholinesterase introduced via liposomes into motor nerve endings block increases in quantal size
.
Pfluegers Arch. Eur. J. Physiol
.
432
,
413
418
.
Brody
,
I.
(
1960
).
The ultrastructure of the tonofibrils in the keratinization process of normal human epidermis
.
J. Ultrastruct. Res
.
4
,
264
297
.
Brown
,
M. R.
and
Chew
,
C. S.
(
1989
).
Carbachol-induced protein phosphorylation in parietal cells: Regulation by intracellular calcium concentration
.
Am J. Physiol
.
257
,
G99
G110
.
Burgoyne
,
R. D.
,
Handel
,
S. E.
,
Morgan
,
A.
,
Rennison
,
M. E.
,
Turner
,
M. D.
and
Wilde
,
C. J.
(
1991
).
Calcium the cytoskeleton and calpactin annexin II in exocytotic secretion from adrenal chromaffin and mammary epithelial cells
.
Biochem. Soc. Trans
19
,
1085
1090
.
Caulfield
,
M. P.
(
1993
).
Muscarinic receptors characterization coupling and function
.
Pharmacol. Ther
.
58
,
319
379
.
Chomczynski
,
P.
and
Sacchi
,
N.
(
1987
).
Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction
.
Anal. Biochem
.
162
,
156
159
.
Dale
,
B. A.
,
Resing
,
K. A.
and
Presland
,
R. B.
(
1994
).
Keratohyalin granule protein
.
In The Keratinocyte Handbook
(ed.
I.
Leigh
,
E. B.
Lane
and
F. M.
Watt
), pp.
323
350
.
Cambridge
:
Cambridge University Press
.
Dale
,
B. A.
,
Presland
,
R. B.
,
Lewis
,
S. P.
,
Underwood
,
R. A.
and
Fleckman
,
P.
(
1997
).
Transient expression of epidermal filaggrin in cultured cells causes collapse of intermediate filament networks with alteration of cell shape and nuclear integrity
.
J. Invest. Dermatol
.
108
,
179
187
.
Delbono
,
O.
,
Gopalakrishnan
,
M.
,
Renganathan
,
M.
,
Monteggia
,
L. M.
,
Messi
,
M. L.
and
Sullivan
,
J. P.
(
1997
).
Activation of the recombinant human α7 nicotinic acetylcholine receptor significantly raises intracellular free calcium
.
J. Pharmacol. Exp. Ther
.
280
,
428
438
.
Derkach
,
V. A.
,
Kurenny
,
D. E.
,
Melishchuk
,
A. I.
,
Selyanko
,
A. A.
and
Skok
,
V. I.
(
1991
).
Role of disulfide bonds in burst-like activity of nicotinic acetylcholine receptor channels in rat sympathetic neurons
.
J. Physiol
.
440
,
1
16
.
DiPalma
,
J. R.
(
1994
).
Basic Pharmacology in Medicine
.
West Chester
:
Medical Surveillance Inc
.
Dolezal
,
V.
,
Lisa
,
V.
and
Tucek
,
S.
(
1997
).
Differential effects of the M-1-M-5 muscarinic acetylcholine receptor subtypes on intracellular calcium and on the incorporation of choline into membrane lipids in genetically modified Chinese hamster ovary cell lines
.
Brain Res. Bull
.
42
,
71
78
.
Eglen
,
R. M.
,
Reddy
,
H.
and
Watson
,
N.
(
1994
).
Selective inactivation of muscarinic receptor subtypes
.
Int. J. Biochem
.
26
,
1357
1368
.
Elgoyhen
,
A. B.
,
Johnson
,
D. S.
,
Boulter
,
J.
,
Vetter
,
D. E.
and
Heinemann
,
S.
(
1994
).
α9: An acetylcholine receptor with novel pharmacological properties expressed in rat cochlear hair cells
.
Cell
79
,
705
715
.
Elias
,
P. M.
(
1996
).
Stratum corneum architecture, metabolic activity and interactivity with subjacent cell layers
.
Exp. Dermatol
.
5
,
191
201
.
Elias
,
P. M.
,
Cullander
,
C.
,
Mauro
,
T.
,
Rassner
,
U.
,
Komuves
,
L.
,
Brown
,
B. E.
and
Menon
,
G. K.
(
1998
).
The secretory granular cell: the outermost granular cell as a specialized secretory cell
.
J. Invest. Dermatol. Symp. Proc
.
3
,
87
100
.
Fleckman
,
P.
,
Dale
,
B. A.
and
Holbrook
,
K. A.
(
1985
).
Profilaggrin, a high-molecular-weight precursor of filaggrin in human epidermis and cultured keratinocytes
.
J. Invest. Dermatol
.
85
,
507
512
.
Gandarillas
,
A.
,
Goldsmith
,
L. A.
,
Gschmeissner
,
S.
,
Leigh
,
I. M.
and
Watt
,
F. M.
(
1999
).
Evidence that apoptosis and terminal differentiation of epidermal keratinocytes are distinct processes
.
Exp. Dermatol
.
8
,
71
79
.
Geneser
,
F.
(
1986
).
Textbook of Histology
.
Copenhagen
:
Munksgaard
.
Godman
,
G. C.
,
Miranda
,
A. F.
,
Deitch
,
A. D.
and
Tanenbaum
,
S. W.
(
1975
).
Action of cytochalasin D on cells of established lines. III. Zeiosis and movements at the cell surface
.
J. Cell Biol
.
64
,
644
667
.
Grando
,
S. A.
,
Cabrera
,
R.
,
Hostager
,
B. S.
,
Bigliardi
,
P. L.
,
Blake
,
J. S.
,
Herron
,
M. J.
,
Dahl
,
M. V.
and
Nelson
,
R. D.
(
1993a
).
Computerized microassay of keratinocyte cell-plastic attachment and proliferation for assessing net stimulatory inhibitory and toxic effects of compounds on nonimmortalized cell lines
.
Skin Pharmacol
.
6
,
135
147
.
Grando
,
S. A.
,
Kist
,
D. A.
,
Qi
,
M.
and
Dahl
,
M. V.
(
1993b
).
Human keratinocytes synthesize, secrete and degrade acetylcholine
.
J. Invest. Dermatol
.
101
,
32
36
.
Grando
,
S. A.
,
Horton
,
R. M.
,
Pereira
,
E. F. R.
,
Diethelm-Okita
,
B. M.
,
George
,
P. M.
,
Albuquerque
,
E. X.
and
Conti-Fine
,
B. M.
(
1995a
).
A nicotinic acetylcholine receptor regulating cell adhesion and motility is expressed in human keratinocytes
.
J. Invest. Dermatol
.
105
,
774
781
.
Grando
,
S. A.
,
Zelickson
,
B. D.
,
Kist
,
D. A.
,
Weinshenker
,
D.
,
Bigliardi
,
P. L.
,
Wendelschafer-Crabb
,
G.
,
Kennedy
,
W. R.
and
Dahl
,
M. V.
(
1995b
).
Keratinocyte muscarinic acetylcholine receptors: immunolocalization and partial characterization
.
J. Invest. Dermatol
.
104
,
95
100
.
Grando
,
S. A.
,
Horton
,
R. M.
,
Mauro
,
T. M.
,
Kist
,
D. A.
,
Lee
,
T. X.
and
Dahl
,
M. V.
(
1996a
).
Activation of keratinocyte nicotinic cholinergic receptors stimulates calcium influx and enhances cell differentiation
.
J. Invest. Dermatol
.
107
,
412
418
.
Grando
,
S. A.
,
Schofield
,
O. M. V.
,
Skubitz
,
A. P. N.
,
Kist
,
D. A.
,
Zelickson
,
B. D.
and
Zachary
,
C. B.
(
1996b
).
Nodular basal cell carcinoma in vivo vs in vitro: Establishment of pure cell cultures, cytomorphologic characteristics, ultrastructure, immunophenotype, biosynthetic activities, and generation of antisera
.
Arch. Dermatol
.
132
,
1185
1193
.
Grando
,
S. A.
(
1997
).
Biological functions of keratinocyte cholinergic receptors
.
J. Invest. Dermatol. Symp. Proc
.
2
,
41
48
.
Grando
,
S. A.
and
Horton
,
R. M.
(
1997
).
The keratinocyte cholinergic system with acetylcholine as an epidermal cytotransmitter
.
Curr. Opin. Dermatol
.
4
,
262
268
.
Grynkiewicz
,
G.
,
Poenie
,
M.
and
Tsien
,
R. Y.
(
1985
).
A new generation of Ca2+ indicators with greatly improved fluorescence properties
.
J. Biol. Chem
.
260
,
3440
3450
.
Haake
,
A. R.
and
Polakowska
,
R. R.
(
1993
).
Cell death by apoptosis in epidermal biology
.
J. Invest. Dermatol
.
101
,
107
112
.
Hall
,
L. L.
,
Bicknell
,
G. R.
,
Primrose
,
L.
,
Pringle
,
J. H.
,
Shaw
,
J. A.
and
Furness
,
P. N.
(
1998
).
Reproducibility in the quantification of mRNA levels by RT-PCR-ELISA and RT competitive PCR-ELISA
.
Biotechniques
24
,
652
656
.
Ikeda
,
K.
,
Ishigaki
,
M.
,
Wu
,
D.
,
Sunose
,
H.
,
Suzuki
,
M.
,
Ishitani
,
K.
and
Takasaka
,
T.
(
1995
).
Intracellular Ca2+ responses induced by acetylcholine in the submucosal nasal gland acinar cells in guinea pigs
.
Am. J. Physiol
.
268
,
L361
L367
.
Ishida-Yamamoto
,
A.
,
Takahashi
,
H.
,
Presland
,
R. B.
,
Dale
,
B. A.
and
Iizuka
,
H.
(
1998
).
Translocation of profilaggrin N-terminal domain into keratinocyte nuclei with fragmented DNA in normal human skin and loricrin keratoderma
.
Lab. Invest
.
78
,
1245
1253
.
Jagger
,
D. J.
,
Griesinger
,
C. B.
,
Rivolta
,
M. N.
,
Holley
,
M. C.
and
Ashmore
,
J. F.
(
2000
).
Calcium signalling mediated by the 9 acetylcholine receptor in a cochlear cell line from the Immortomouse
.
J. Physiol
.
527
,
49
54
.
Jewell
,
S. A.
,
Bellomo
,
G.
,
Thor
,
H.
,
Orrenius
,
S.
and
Smith
,
M.
(
1982
).
Bleb formation in hepatocytes during drug metabolism is caused by disturbances in thiol and calcium ion homeostasis
.
Science
217
,
1257
1259
.
Jones
,
S. V. P.
(
1993
).
Muscarinic receptor subtypes modulation of ion channels
.
Life Sci
.
52
,
457
464
.
Kebabian
,
J. W.
and
Neumeyer
,
J. L.
, editors
(
1994
).
The RBI Handbook of Receptor Classification
.
Natick
:
Research Biochemicals International
.
Kerr
,
J. F.
,
Wyllie
,
A. H.
and
Currie
,
A. R.
(
1972
).
Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics
.
Br. J. Cancer
26
,
239
257
.
Klapproth
,
H.
,
Reinheimer
,
T.
,
Metzen
,
J.
,
Munch
,
M.
,
Bittinger
,
F.
,
Kirkpatrick
,
C. J.
,
Hohle
,
K. D.
,
Schemann
,
M.
,
Racke
,
K.
and
Wessler
,
I.
(
1997
).
Non-neuronal acetylcholine, a signalling molecule synthezised by surface cells of rat and man
.
Naunyn-Schmiedebergs Arch. Pharmacol
.
355
,
515
523
.
Krawczyk
,
W. S.
and
Wilgram
,
G. F.
(
1973
).
Hemidesmosome and desmosome morphogenesis during epidermal wound healing
.
J. Ultrastruct. Res
.
45
,
93
101
.
Kurosumi
,
K.
and
Kawabata
,
I.
(
1976
).
Transmission and scanning electron microscopy of the human ceruminous apocrine gland. I. Secretory glandular cells
.
Arch. Histol. Jpn
39
,
207
229
.
Kwon
,
O. S.
,
Chung
,
J. H.
,
Cho
,
K. H.
,
Suh
,
D. H.
,
Park
,
K. C.
,
Kim
,
K. H.
and
Eun
,
H. E.
(
1999
).
Nicotine-enhanced epithelial differentiation in reconstructed human oral mucosa in vitro
.
Skin Pharmacol. Appl. Skin Physiol
.
12
,
227
234
.
Lavker
,
R. M.
and
Matoltsy
,
A. G.
(
1970
).
Formation of horny cells: the fate of cell organelles and differentiation products in ruminal epithelium
.
J. Cell Biol
.
44
,
501
512
.
Lazareno
,
S.
and
Birdsall
,
N. J.
(
1993
).
Pharmacological characterization of acetylcholine-stimulated [35S]-GTP gamma S binding mediated by human muscarinic m1-m4 receptors: antagonist studies
.
Br. J. Pharmacol
.
109
,
1120
1127
.
Lee
,
Y.-S.
,
Yuspa
,
S. H.
and
Dlugosz
,
A. A.
(
1998
).
Differentiation of cultured human epidermal keratinocytes at high cell densities is mediated by endogenous activation of the protein kinase C signaling pathway
.
J. Invest. Dermatol
.
11
,
762
766
.
LeSage
,
G.
,
Alvaro
,
D.
,
Benedetti
,
A.
,
Glaser
,
S.
,
Marucci
,
L.
,
Baiocchi
,
L.
,
Eisel
,
W.
,
Caligiuri
,
A.
,
Phinizy
,
J. L.
,
Rodgers
,
R.
et al.  (
1999
).
Cholinergic system modulates growth, apoptosis, and secretion of cholangiocytes from bile duct-ligated rats
.
Gastroenterology
117
,
191
199
.
Lewis
,
J. E.
,
Jensen
,
P. J.
and
Wheelock
,
M. J.
(
1994
).
Cadherin function is required for human keratinocytes to assemble desmosomes and stratify in response to calcium
.
J. Invest. Dermatol
.
102
,
870
877
.
Li
,
L.
,
Tucker
,
R. W.
,
Hennings
,
H.
and
Yuspa
,
S. H.
(
1995
).
Chelation of intracellular Ca2+ inhibits murine keratinocyte differentiation in vitro
.
J. Cell Physiol
.
163
,
105
114
.
Majno
,
G.
and
Joris
,
I.
(
1995
).
Apoptosis, oncosis, and necrosis: An overview of cell death
.
Am. J. Pathol
.
146
,
3
15
.
Matoltsy
,
A. G.
(
1966
).
Membrane-coating granules of the epidermis
.
J. Ultrastruct. Res
.
15
,
510
515
.
Mauro
,
T.
,
Bench
,
G.
,
Sidderas-Haddad
,
E.
,
Feingold
,
K.
,
Elias
,
P.
and
Cullander
,
C.
(
1998
).
Acute barrier perturbation abolishes the Ca2+ and K+ gradients in murine epidermis: quantitative measurement using PIXE
.
J. Invest. Dermatol
.
111
,
1198
1201
.
McCall
,
C. A.
and
Cohen
,
J. J.
(
1991
).
Programmed cell death in terminally differentiating keratinocytes: role of endogenous endonuclease
.
J. Invest. Dermatol
.
97
,
111
114
.
Menon
,
G. K.
,
Brown
,
B. E.
and
Elias
,
P. M.
(
1986
).
Avian epidermal differentiation: Role of lipids in permeability barrier formation
.
Tissue & Cell
18
,
71
82
.
Menon
,
G. K.
,
Elias
,
P. M.
,
Lee
,
S. H.
and
Feingold
,
K. R.
(
1992
).
Localization of calcium in murine epidermis following disruption and repair of the permeability barrier
.
Cell Tiss. Res
.
270
,
503
512
.
Menon
,
G. K.
,
Price
,
L. F.
,
Bommannan
,
B.
,
Elias
,
P. M.
and
Feingold
,
K. R.
(
1994
).
Selective obliteration of the epidermal calcium gradient leads to enhanced lamellar body secretion
.
J. Invest. Dermatol
.
102
,
789
795
.
Meyer
,
J.
,
Alvares
,
O. F.
and
Barrington
,
E. P.
(
1970
).
Volume and dry weight of cells in the epithelium of rat cheek and palate
.
Growth
34
,
57
73
.
Ndoye
,
A.
,
Buchli
,
R.
,
Greenberg
,
B.
,
Nguyen
,
V. T.
,
Zia
,
S.
,
Rodriguez
,
J. G.
,
Webber
,
R. J.
,
Lawry
,
M. A.
and
Grando
,
S. A.
(
1998
).
Identification and mapping of keratinocyte muscarinic acetylcholine receptor subtypes in human epidermis
.
J. Invest. Dermatol
.
111
,
100
106
.
Nguyen
,
V. T.
,
Lee
,
T. X.
,
Ndoye
,
A.
,
Shultz
,
L. D.
,
Pittelkow
,
M. R.
,
Dahl
,
M. V.
,
Lynch
,
P. J.
and
Grando
,
S. A.
(
1998
).
The pathophysiological significance of non-desmoglein targets of pemphigus autoimmunity. Pemphigus vulgaris and foliaceus patients develop antibodies against keratinocyte cholinergic receptors
.
Arch. Dermatol
.
134
,
971
980
.
Nguyen
,
V. T.
,
Hall
,
L. L.
,
Gallacher
,
G.
,
Ndoye
,
A.
,
Jolkovsky
,
D. L.
,
Webber
,
R. J.
,
Buchli
,
R.
and
Grando
,
S. A.
(
2000a
).
Choline acetyltransferase, acetylcholinesterase, and nicotinic acetylcholine receptors of human gingival and esophageal epithelia
.
J. Dental Res
.
79
,
939
949
.
Nguyen
,
V. T.
,
Ndoye
,
A.
and
Grando
,
S. A.
(
2000b
).
Novel human α9 acetylcholine receptor regulating keratinocyte adhesion is targeted by pemphigus vulgaris autoimmunity
.
Am. J. Pathol
.
157
,
1377
1391
.
Odland
,
G. F.
(
1971
).
Histology and fine structure of the epidermis
.
In The Skin
(ed.
E. B.
Helwig
and
F. K.
Mostofi
), pp.
28
46
.
Baltimore
:
Williams & Wilnins Company
.
Odland
,
G. P.
and
Holbrook
,
K.
(
1987
).
The lamellar granules of the epidermis
.
Curr. Probl. Dermatol
.
9
,
29
49
.
Odland
,
G. F.
(
1991
).
Structure of the skin
.
In Physiology, Biochemistry, and Molecular Biology of the Skin
(ed.
L. A.
Goldsmith
), pp.
3
62
.
New York
:
Oxford University Press
.
Orrenius
,
S.
and
Nicotera
,
P.
(
1994
).
The calcium and cell death
.
J. Neur. Transm. (suppl
.)
43
,
1
11
.
Perrin
,
D.
,
Moeller
,
K.
,
Hanke
,
K.
and
Soling
,
H. D.
(
1992
).
cAMP and calcium-mediated secretion in parotid acinar cells is associated with reversible changes in the organization of the cytoskeleton
.
J. Cell Biol
.
116
,
127
134
.
Puchelle
,
E.
,
Beorchia
,
A.
,
Menager
,
M.
,
Zahm
,
J. M.
and
Ploton
,
D.
(
1991
).
Three-dimensional imaging of the mucus secretory process in the cryofixed frog respiratory epithelium
.
Biol. Cell
72
,
159
166
.
Rawlings
,
A. V.
,
Scott
,
I. R.
,
Harding
,
C. R.
and
Bowser
,
P. A.
(
1994
).
Stratum corneum moisturization at the molecular level
.
J. Invest. Dermatol
.
103
,
731
741
.
Rinner
,
I.
,
Globerson
,
A.
,
Kawashima
,
K.
,
Korsatko
,
W.
and
Schauenstein
,
K.
(
1999
).
A possible role for acetylcholine in the dialogue between thymocytes and thymic stroma
.
Neuroimmunomodulation
6
,
51
55
.
Sastry
,
B. V.
and
Sadavongvivad
,
C.
(
1978
).
Cholinergic systems in non-nervous tissues
.
Pharmacol. Rev
.
30
,
65
132
.
Scott
,
I. R.
and
Harding
,
C. R.
(
1981
).
Studies on the synthesis and degradation of a high molecular weight, histidine-rich phosphoprotein from mammalian epidermis
.
Biochim. Biophys. Acta
669
,
65
78
.
Scott
,
I. R.
and
Harding
,
C. R.
(
1986
).
Filaggrin breakdown to water binding compounds during development of the rat stratum corneum is controlled by the water activity of the environment
.
Dev. Biol
.
115
,
84
92
.
Seguela
,
P.
,
Wadiche
,
J.
,
Dineley-Miller
,
K.
,
Dani
,
J. A.
and
Patrick
,
J. W.
(
1993
).
Molecular cloning, functional properties, and distribution of rat brain α7: a nicotinic cation channel highly permeable to calcium
.
J. Neurosci
.
13
,
596
604
.
Sharpe
,
G. R.
,
Fisher
,
C.
,
Gillespie
,
J. I.
and
Greenwell
,
J. R.
(
1993
).
Growth and differentiation stimuli induce different and distinct increases in intracellular free calcium in human keratinocytes
.
Arch Dermatol. Res
.
284
,
445
450
.
Smith
,
J. B.
and
Fenske
,
N. A.
(
1996
).
Cutaneous manifestations and consequences of smoking
.
J. Am. Acad. Dermatol
.
34
,
717
732
.
Soltoff
,
S. P.
and
Toker
,
A.
(
1995
).
Carbachol, substance P, and phorbol ester promote the tyrosine phosphorylation of protein kinase C delta in salivary gland epithelial cells
.
J. Biol. Chem
.
270
,
13490
13495
.
Theilig
,
C.
,
Bernd
,
A.
,
Ramirez-Bosca
,
A.
,
Gormar
,
F. F.
,
Bereiter-Hahn
,
J.
,
Keller-Stanislawski
,
B.
,
Sewell
,
A. C.
,
Rietbrock
,
N.
and
Holzmann
,
H.
(
1994
).
Reactions of human keratinocytes in vitro after application of nicotine
.
Skin Pharmacol
.
7
,
307
315
.
Thody
,
A. J.
and
Shuster
,
S.
(
1989
).
Control and function of sebaceous glands
.
Physiol. Rev
.
69
,
383
416
.
Verbitsky
,
M.
,
Rothlin
,
C. V.
,
Katz
,
E.
and
Belen Elgoyhen
,
A.
(
2000
).
Mixed nicotinic-muscarinic properties of the α9 nicotinic cholinergic receptor
.
Neuropharmacology
39
,
2515
2524
.
Wessler
,
I.
,
Kirkpatrick
,
C. J.
and
Racke
,
K.
(
1998
).
Non-neuronal acetylcholine, a locally acting molecule, widely distributed in biological systems: expression and function in humans
.
Pharmacol. Ther
.
77
,
59
79
.
Wessler
,
I.
,
Kirkpatrick
,
C. J.
and
Racke
,
K.
(
1999
).
The cholinergic ‘pitfall’: acetylcholine, a universal cell molecule in biological systems, including humans
.
Clin. Exp. Pharmacol. Physiol
.
26
,
198
205
.
Wikstrom
,
M. A.
,
Lawoko
,
G.
and
Heilbronn
,
E.
(
1998
).
Cholinergic modulation of extracellular ATP-induced cytoplasmic calcium concentrations in cochlear outer hair cells
.
J. Physiol. Paris
92
,
345
349
.
Williams
,
A. C.
,
Cornwell
,
P. A.
and
Barry
,
B. W.
(
1992
).
On the non-Gaussian distribution of human skin permeabilities
.
Int. J. Pharm
.
86
,
69
77
.
Wolff
,
K.
and
Schreiner
,
E.
(
1968
).
An electron microscopic study on the extraneous coat of keratinocytes and the intercellular space of the epidermis
.
J. Invest. Dermatol
.
51
,
418
430
.
Yan
,
G.-M.
,
Lin
,
S.-Z.
,
Irwin
,
R. P.
and
Paul
,
S. M.
(
1995
).
Activation of muscarinic cholinergic receptors blocks apoptosis of cultured cerebellar granule neurons
.
Mol. Pharmacol
.
47
,
248
257
.
Yang
,
X.
and
Buccafusco
,
J. J.
(
1994
).
Effect of chronic central treatment with the acetylcholine analog methylcarbamylcholine on cortical nicotinic receptors: Correlation between receptor changes and behavioral function
.
J. Pharmacol. Exp. Ther
.
271
,
651
659
.
Zia
,
S.
,
Ndoye
,
A.
,
Nguyen
,
V. T.
and
Grando
,
S. A.
(
1997
).
Nicotine enhances expression of the α3, α4, α5, and α7 nicotinic receptors modulating calcium metabolism and regulating adhesion and motility of respiratory epithelial cells
.
Res. Commun. Mol. Pathol. Pharmacol
.
97
,
243
262
.
Zia
,
S.
,
Ndoye
,
A.
,
Lee
,
T. X.
,
Webber
,
R. J.
and
Grando
,
S. A.
(
2000
).
Receptor-mediated inhibition of keratinocyte migration by nicotine involves modulations of calcium influx and intracellular concentration
.
J. Pharm. Exp. Ther
.
293
,
973
981
.