INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE MICROGRAVITY ENVIRONMENT of spaceflight alters many physiological functions in astronauts, and in vitro experiments have suggested that microgravity may bring about some physiological alterations by influencing the activation, distribution, and translocation of protein kinase C (PKC) (5, 17, 22, 27, 30). PKC was recently shown to modulate Ca²⁺-dependent signal transduction in cultured ciliated airway epithelial cells (29). We hypothesized that microgravity may influence PKC-modulated Ca²⁺ signaling. We have adapted an Earth-based culture system that simulates microgravity to study possible effects of reduced gravitational force on Ca²⁺ signaling in these cells (11).

Two signaling pathways, one initiated by mechanical stimulation and the other by purinergic receptor activation via ATP addition (7), were shown to be modulated by PKC (29). Both lead to an increase in ciliary beat frequency in airway epithelial cells (10, 25). Mechanical stimulation of a single cell causes cell-to-cell spread of increased intracellular Ca²⁺ concentration ([Ca²⁺]_i), referred to as a Ca²⁺ "wave" (24, 29). In airway epithelial cultures, Ca²⁺ waves appear to be propagated by diffusion of inositol trisphosphate (IP₃) through gap junctions (2, 24). Bath application of ATP generates dampened intracellular Ca²⁺ oscillations, at a frequency of up to two per minute. Cells responding to ATP appear to act as asynchronous, autonomous units, without communicating the Ca²⁺ signal to adjacent cells (10, 15, 29).

Both mechanical and ATP stimulation activate phospholipase C (PLC) generation of IP₃ and IP₃-dependent Ca²⁺ mobilization to increase [Ca²⁺]_i (10, 12, 16). Concomitant PLC generation of diacylglycerol and diacylglycerol-dependent activation of PKC appear to be important in shaping the cellular responses. Pharmacological activation of PKC before mechanical stimulation limits the extent of the Ca²⁺ wave, suggesting that PKC may act as an inhibitor of gap junctional Ca²⁺ communication. However, PKC activation inhibits ATP- induced Ca²⁺ oscillations that do not apparently involve gap junctions, suggesting that PKC has multiple sites of action that impinge on Ca²⁺ signaling. PKC activation before ATP addition reduces the Ca²⁺ oscillations so that there is a single, reduced Ca²⁺ transient or no [Ca²⁺]_i increase at all. Prior inhibition of PKC causes the ATP response to form a long-lasting [Ca²⁺]_i increase without oscillations. These results are consistent with activated PKC as a feedback inhibitor of ATP-dependent PLC activation and therefore as the generator of the oscillation pattern (29).

Previously, we demonstrated that cells of intact tracheal epithelial explants retain the integrity of their mechanically stimulated intercellular Ca²⁺ signaling after being kept in simulated microgravity (sµG) (11). Because PKC modulates mechanically induced Ca²⁺ waves differently than ATP-induced Ca²⁺ oscillations, we extended our previous work on the effects of sµG to examine both signaling pathways to increase our chances of revealing potential effects of microgravity. Also, to improve our chances of detecting possible effects of microgravity, we have developed a thinner preparation of intact epithelial sheets that would allow transmission of a stronger fluorescent Ca²⁺ signal than the thick explants used previously (11). In addition, tracheal epithelial outgrowths cultured on semipermeable membranes in dual-chambered bioreactors were assayed to relate results of this study to those characterized previously in tracheal epithelial outgrowths cultured on glass coverslips (2, 16, 24, 25, 29). To simulate microgravity in the laboratory, outgrowth cultures and epithelial sheets were rotated in bioreactors with the axis of rotation perpendicular to the Earth's gravity so that cells experienced a constantly changing gravity vector. Control cultures were rotated around an axis parallel to gravity for the unit gravity (G) condition. In this report, we show that the effects of pharmacological PKC activation on both mechanically induced and ATP-induced Ca²⁺ signaling are more robust in airway epithelial cells kept in sµG, suggesting an increase in PKC sensitivity in cells kept in microgravity.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Outgrowth culture preparation. Tracheal mucosae were dissected from New Zealand White rabbits, and rat tail collagen was prepared as previously described (6). After the mucosa was cut into ~0.5-mm² pieces, the explants were plated on collagen-coated semipermeable Nucleopore membranes (Corning, Acton, MA), which had been glued with silicone rubber in a six-well disk. Culture medium consisting of DMEM supplemented with 25 mM HEPES (pH 7.4), 10% fetal bovine serum, 100 U/ml penicillin, 100 µg/ml streptomycin, 0.25 µg/ml amphotericin B, and 0.37% (wt/vol) NaHCO₃, referred to as sDMEM (all culture reagents purchased from GIBCO BRL, Rockville, MD), was added to barely cover the membranes, keeping the cells at the air-liquid interface as they are in situ. After 1 wk of explant attachment and cell growth, the six-well disk was assembled within a low-sheer-stress dual-chambered bioreactor (Synthecon, Houston, TX) with the apical cell surfaces facing the air chamber and the basolateral surfaces facing the culture medium chamber (Fig. 1A). The medium chamber was completely filled with warmed sDMEM. Humidified 5% CO₂-95% air was pumped through the core oxygenator silicone membrane into both chambers. Bubbles were removed, and the sDMEM volume was adjusted so that it was at an equilibrium pressure with the warmed air chamber. Outgrowth cultures were rotated at about 10 rpm in the bioreactor for 4-10 days. The axis of rotation was vertical in unit G (Fig. 1A) or horizontal in sµG (Fig. 1B), in which cells were subjected to a cyclically changing gravity vector and experienced a time-averaged gravitational force of zero. To prepare for Ca²⁺ imaging, cultures were incubated in 5 µM fura 2-AM (Molecular Probes, Eugene, OR), a Ca²⁺-sensitive fluorescent dye, in Hanks' balanced salt solution supplemented with 25 mM HEPES (pH 7.2), referred to as sHBSS, for 45 min at 37°C in the dark (as described in Refs. 24 and 29).

View larger version (186K): [in this window] [in a new window]   Fig. 1.   Assembly diagram (A) and photograph (B) of a dual-chambered bioreactor. Apical cell surfaces of outgrowth tracheal epithelial cultures were positioned to face the air chamber, with the semipermeable membranes facing the culture medium chamber. As shown in B, for the simulated microgravity (sµG) condition, a motor continuously rotated the dual-chambered bioreactor around a horizontal axis so that the orientation of the outgrowth cells relative to gravity was constantly changing. (For the control, not shown, the bioreactor was rotated around a vertical axis to maintain unit G.)

Epithelial sheet preparation. Tracheal mucosae were positioned apical side up on 0.2-µm-pore size Supor-200 membranes (Gelman Sciences, Ann Arbor, MI) overlying a pool of 40 U/ml collagenase IIIA (Sigma Chemical, St. Louis, MO) in DMEM with 25 mM HEPES for 18 h at 4°C. After the collagenase treatment, intact sheets of epithelium were microdissected free from the underlying connective tissue. Suspensions of epithelial sheets in sDMEM were incubated with 1% fetal bovine serum in untreated culture dishes for the first 1 or 2 days. For the sµG condition, sheets were rotated for 3-6 days at ~20 rpm within a zero-head space, single-chambered HARV bioreactor (Synthecon). Epithelial sheets and medium rotated with the vessel, remaining freely suspended and not colliding with the vessel wall. The epithelial sheets experienced randomly changing gravity and possibly less tension than outgrowth cultures because they were not anchored to a rigid substrate. Bubbles were removed daily to minimize turbulence and sheer. Epithelial sheets were loaded with fura 2 by exposing them to 10 µM fura 2-AM for 2 h at 37°C and then were adhered to coverslips freshly coated with a thin film of Dermabond (Ethicon, Somerville, NJ) before proceeding with Ca²⁺ imaging. Tissue adhesion using Dermabond, a 2-octyl cyanoacrylate glue, did not involve integrin binding between the cytoskeleton and substrate.

[Ca²⁺]_i measurement. Fura 2 fluorescence image analysis was conducted as previously detailed (24, 21). In brief, fura 2-loaded cells incubating in sHBSS on glass coverslips were mounted over an inverted Nikon (Garden City, NY) Diaphot microscope. Cells were viewed through a ×40 fluor 1.3-numerical aperture oil-immersion objective and quartz optics. A 100-W mercury lamp provided excitation light, filtered primarily through a 380-nm bandpass filter or alternately through a 340-nm filter every 15 s, and then passed through a 405-nm dichroic mirror. Long-pass >510-nm emission light was captured by a silicon-intensified target camera (Cohu, San Diego, CA) and then recorded by an optical memory disk recorder (Panasonic, Secaucus, NJ). A frame grabber and image processor boards (Data Translation, Marlborough, MA) digitized images for analyses. Data acquisition and analysis software designed by Dr. Michael J. Sanderson was used to perform background subtraction, shading correction, calibration, and ratiometric calculation of [Ca²⁺]_i [using the formulas of Grynkiewicz et al. (13)] for a four × four pixel area in the middle of each cell. Images were recorded at one to two frames per second.

Mechanical stimulation. A ciliated cell was mechanically stimulated by touching the apical cell surface with the tip of a fire-polished micropipette for ~150 ms with the use of a piezoelectric device and hydraulic micromanipulator (Narishige). Cell viability after mechanical stimulation was assessed by ensuring that cilia were still beating and that fura 2 dye was not lost (i.e., that fluorescence was not low at both 340 and 380 nm). Experimental trials resulting in an injured stimulated cell or no Ca²⁺ wave (presumably because the deflected micropipette tip did not sufficiently contact the cell) were not included.

Data presentation. Quantitative data are reported as means ± SE with n equal to the number of cells, except where noted otherwise. Differences were considered significant when P < 0.05 by the Student's t-test. Only [Ca²⁺]_i increases >25 nM were sizable enough to be considered significant. Occasionally, [Ca²⁺]_i values in a small area of a microscopic field could not be analyzed because of uneven focus, debris, or excessive thickness. Cells within such areas were not included in computations of mean [Ca²⁺]_i.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cells cultured at an air-liquid interface on a semipermeable substrate show a thicker, more differentiated morphology. When tracheal epithelial explants were cultured at the air-liquid interface on semipermeable membranes (Fig. 1), the outgrowths from those explants had a higher proportion of ciliated cells compared with our previously used immersed cultures grown on collagen-coated glass coverslips (2, 16, 24, 29). Also, cross sections of fixed outgrowth cells cultured on semipermeable membranes revealed a squamous to cuboidal morphology, taller and more cuboidal than outgrowth cultures grown on glass coverslips.

PKC activation inhibited mechanically stimulated Ca²⁺ waves more markedly in outgrowth cultures kept at sµG than at unit G. Figure 2A shows a typical Ca²⁺ wave that spreads radially from the mechanically stimulated cell to over 40 surrounding cells within 10 s. Figure 2B shows an example of the smaller mechanically induced Ca²⁺ wave that results after exposure to the PKC activator 12-O-tetradecanoylphorbol-13-acetate (TPA). Under control conditions (no TPA added), mechanical stimulation evoked similar intercellular Ca²⁺ waves in cultures maintained under both unit G and sµG conditions. Increased [Ca²⁺]_i propagated among a mean 35 ± 4 cells (n = 52) in unit G control cultures, similar to the mean 38 ± 5 cells (n = 44) in outgrowth cultures kept in sµG (Fig. 3A). Prior activation of PKC by 5- to 20-min treatment with 80 nM TPA diminished the extent of cell-to-cell spread of increased [Ca²⁺]_i to 10 ± 2 (n = 17) cells in unit G cultures. The inhibitory effect of TPA was slightly greater in sµG cultures, in which the spread of Ca²⁺ waves was limited to only 7 ± 1 cells (n = 25) (significantly different at P = 0.002; Fig. 3A). Exposure to a TPA analog that does not activate PKC, 4-phorbol-12,13-didecanoate, did not alter mechanically induced Ca²⁺ waves [37 ± 8 responding cells (n = 11) in unit G cultures and 38 ± 7 responding cells (n = 9) in sµG cultures].

View larger version (91K): [in this window] [in a new window]   Fig. 2.   A: mechanical stimulation of a single cell (at arrow) caused intracellular Ca2+ concentration ([Ca2+]i) to first increase in the stimulated cell and then spread radially to adjacent cells. Time series shown is a representative example from a unit G culture; similar results were observed in cells kept in sµG. Nanomolar [Ca2+]i is portrayed in pseudocolor (see color bar). B: prior protein kinase C (PKC) activation [80 nM 12-O-tetradecanoylphorbol-13-acetate (TPA) for 5-20 min] restricted the extent of spread of the intercellular Ca2+ wave.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   A: number of cells participating in a mechanically stimulated Ca2+ wave in unit G cultures was roughly equivalent to the number in sµG cultures under control conditions. Extent of spread of the Ca2+ wave was diminished by TPA but not altered by the inactive analog, 4-phorbol-12,13-didecanoate (4-phorbol). B and C: average control [Ca2+]i magnitudes observed in mechanically stimulated (MS) cells (solid lines) were larger than in secondary (2°) cells (dash-dot lines). Ca2+ responses of MS and 2° cells in unit G outgrowth cultures (B) were nearly the same as those from sµG cultures (C). PKC activation significantly reduced the mean [Ca2+]i response in 2° cells (dotted lines) of sµG cultures (C) to a greater extent than in 2° cells of unit G cultures (B). Small reductions in the [Ca2+]i amplitudes of MS cells treated with TPA (dashed lines) were not significant.

Reduction of the amplitude and number of ATP-induced Ca²+ oscillations in outgrowth cultures by PKC activation. ATP addition initiated Ca²⁺ oscillations in outgrowth cultures from both unit G and sµG. A greater proportion of cells oscillated and the frequency of oscillation increased with higher ATP concentrations of 1-4 µM (Fig. 4) in cultures from both gravity conditions. A subsaturating ATP concentration (2 µM) was used to generate oscillations in experiments in which the effects of sµG and the effects of TPA addition were going to be assessed. Most cells generate at least one oscillation in response to 2 µM ATP (see below); this first Ca²⁺ oscillation provides a convenient measure of magnitude. The maximum [Ca²⁺]_i of the first oscillation in unit G cultures was 216 ± 11 nM (n = 71; Fig. 5A). The first oscillation in cultures from sµG rose to 241 ± 11 nM Ca²⁺ (n = 78; Fig. 5B), a small but significant increase over the unit G condition. Activating PKC (~25 min, 160 nM TPA) before ATP application caused a significant 29% reduction in the magnitude of the initial [Ca²⁺]_i increase in sµG cultures (to 172 ± 37 nM, n = 21), whereas the 14% reduction in initial [Ca²⁺]_i elevation in unit G cultures (to 186 ± 31 nM, n = 22) was not significant. The ATP-induced [Ca²⁺]_i increases observed in cultures kept at unit G or in sµG were indistinguishable in rates of increase in [Ca²⁺]_i and recovery to resting [Ca²⁺]_i.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Representative traces from individual cells at the indicated ATP concentration ([ATP]) show that the numbers of oscillations per minute decreased with lower [ATP] in both unit G and sµG cultures.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Average magnitude of the first [Ca2+]i increase in response to 2 µM ATP (added at ~20 s) was slightly lower in unit G cells (A) than in sµG cells (B). PKC activation significantly reduced the magnitude of the ATP-induced Ca2+ transient in sµG cultures and curtailed subsequent oscillations in both unit G and sµG cultures. Distribution of the number of oscillations exhibited by individual cells within 220 s of exposure to 2 µM ATP is shown for both unit G (C) and sµG (D) outgrowth cultures. Number of cells responding to ATP with one or more Ca2+ oscillations declined after treatment with TPA to activate PKC (). Ca2+ oscillations in cultures treated with 4-phorbol, a TPA analog (), resembled those in untreated cultures ().

Comparable morphology of cells in epithelial sheets at unit G and sµG. To further resolve the effects of sµG on Ca²⁺ signaling and PKC activation, we attempted to also simulate microgravity by suspending epithelial sheets within a rotating HARV bioreactor. Unlike outgrowth cultures, epithelial sheets are not attached to a substrate, and suspended epithelial sheets gradually tumble in their orbits within the HARV bioreactor; thus they experience a randomized gravity vector, differing from outgrowth cultures subjected to a cyclically changing gravity vector in the dual-chambered bioreactor. Epithelial sheets contract after being microdissected free of their underlying connective tissue. When viewed in cross section, epithelial sheets are more highly ciliated and cells have a more columnar shape than outgrowth culture cells. Unit G epithelial sheet cells had a columnar morphology with a mean height-to-width ratio of 1.44 ± 0.06 (n = 48 cells), whereas sµG epithelial sheet cells had a height-to-width ratio of 1.45 ± 0.06 (n = 42 cells).

PKC activation inhibited mechanically induced Ca²+ waves in intact epithelial sheets kept at sµG but not sheets kept at unit G. Mechanical stimulation of a single cell near the edge of an intact tracheal epithelial sheet caused cell-to-cell propagation of increased [Ca²⁺]_i away from the site of stimulation, which could be observed primarily along the periphery of the epithelial sheet (Fig. 6). Cell-to-cell spread of a Ca²⁺ wave into the middle of the epithelial sheet could be discerned in a minority of records when the epithelial sheet laid flat against the coverslip without any folding. An intercellular Ca²⁺ wave propagated through a radius of 3.8 ± 0.3 (n = 31) cells in control unit G epithelial sheets (Fig. 7A). This result was not significantly different from the Ca²⁺ wave radius of 4.4 ± 0.4 cells (n = 21) in epithelial sheets maintained in sµG.

View larger version (67K): [in this window] [in a new window]   Fig. 6.   Mechanical stimulation (at arrow) initiated a Ca2+ wave in an intact epithelial sheet. Comparable results were observed in control tissues maintained in sµG and in TPA-treated tissues maintained at unit G. Spread of Ca2+ waves in TPA-treated epithelial sheets kept in sµG was slightly less (not shown). Nanomolar [Ca2+]i is represented in pseudocolor (see color bar).

View larger version (20K): [in this window] [in a new window]   Fig. 7.   A: mean radius of spread of mechanically induced Ca2+ waves in unit G epithelial sheets was not significantly different from that in sµG epithelial sheets under control conditions. Unlike the results in outgrowth cultures (Fig. 3A), PKC activation (10-45 min of 160 nM TPA) did not significantly inhibit the mean radius of spread in epithelial sheets kept at unit G and only slightly depressed the spread of Ca2+ waves in epithelial sheets kept in sµG. B: in unit G epithelial sheets, the Ca2+ transients in both MS cells (solid lines) and 2° cells (dash-dot lines) were roughly the same before and after exposure to TPA (dashed and dotted lines). C: in sµG epithelial sheets, TPA treatment reduced the amplitude of the [Ca2+]i increases of both MS and 2° cells.

Magnitude and frequency of ATP-induced Ca²+ oscillations in epithelial sheets reduced by activation of PKC. Epithelial sheets required an ATP concentration of 5 µM to stimulate a similar proportion of responding cells as seen in outgrowth cultures stimulated by 2 µM ATP. The Ca²⁺ oscillation frequency increased and peak [Ca²⁺]_i increased from 2.5 to 10 µM. At higher ATP concentrations, oscillations were suppressed and a single, longer-lasting [Ca²⁺]_i increase was observed (not shown). The net [Ca²⁺]_i amplitudes induced by 5 µM ATP were ~50 nM greater in epithelial sheets than the [Ca²⁺]_i amplitudes induced by 2 µM ATP in outgrowth cultures.

View larger version (22K): [in this window] [in a new window]   Fig. 8.   A: in epithelial sheets kept at unit G, the average Ca2+ response to 5 µM ATP (solid line) was attenuated after PKC activation with TPA (dashed line). B: similarly, in epithelial sheets kept in sµG, TPA-treated cells (solid line) showed a diminished Ca2+ response to 5 µM ATP compared with controls (dashed line). C and D: histogram distribution of the number of Ca2+ oscillations exhibited by cells within 220 s of 5 µM ATP addition at unit G (C) and at sµG (D). Proportion of cells exhibiting no oscillations increased after TPA treatment () in unit G epithelial sheets (C) and in sµG epithelial sheets (D), compared with controls ().

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The effects of TPA addition observed here in the outgrowth cultures of airway epithelial cells under the control incubation condition (unit G) are similar to the results obtained previously (29). The goal of these experiments was to test the hypothesis that microgravity alters PKC-modulated Ca²⁺ signaling in tracheal epithelial cells. The principal effect of sµG was an increase in the sensitivity of cells to TPA. This is apparent in the Ca²⁺ responses to both mechanical stimulation (see Table 1) and ATP addition (Figs. 5, A and B, and 8, A and B). An increase in TPA sensitivity could occur if more PKC was available for TPA-induced activation. Nascent, inactive PKC associates with the cytoskeleton. It becomes phosphorylated to form mature PKC and then translocates to the cytosol or to anchoring proteins on the plasma membrane before it can be activated by TPA or its endogenous activator, diacylglycerol (23). sµG may cause an "upregulation" of PKC by generally increasing the copy number in the cell or by making more mature PKC available to TPA-induced activation.

Different studies have proposed divergent effects of microgravity on PKC, some suggesting that PKC activity is increased and others that PKC activity is decreased. For instance, microgravity during spaceflight reduced tumor necrosis factor-stimulated lysis of LM929 cells. This lesser cytotoxic effect of tumor necrosis factor in microgravity was restored to the unit G level by inhibition of PKC (30), suggestive of higher PKC activity in microgravity. In U-937 myelomonocytic and Jurkat T cells, the microgravity environment of space decreased the cytosolic fraction, increased the nuclear fraction distribution of PKC (27), and changed the kinetics of PKC translocation on activation with a phorbol ester related to TPA (17). On the other hand, the microgravity environment in a sounding rocket decreased c-fos and c-jun mRNA expression induced by TPA or epidermal growth factor, an agonist of a PKC-mediated pathway, in epidermoid carcinoma cells (5), suggesting that microgravity inhibits the PKC activation pathway. In addition, the weightlessness of space inhibited interleukin production induced by TPA in leukocytes, suggesting that a PKC-mediated signaling pathway is inhibited in microgravity (22).

PKC is associated with numerous cytoskeletal components (20). Mechanotransduction may begin with physical distortion of the cytoskeletal architecture that, in turn, is transduced to biochemical changes (1, 18). Cytoskeletal tension is generated by focal adhesion complex attachments between the cytoskeleton and substrate as well as by junctional complexes between adjoining cells (18). The lack of gravity in any one direction for a sufficient duration could reduce cytoskeletal tension by disrupting the preexisting balance of forces distributed between adhesion complexes and the cytoskeleton (19). The cytoskeleton may thereby sense microgravity (3, 19) and influence PKC.

Although the focus of these experiments was to assay the effects of sµG, the decision to use epithelial sheets as well as outgrowth cultures gave a surprising result independent of sµG. In outgrowth cultures, PKC activation by TPA treatment dramatically restricted the spread of mechanically stimulated Ca²⁺ waves, and this effect was very robust. However, TPA did not inhibit the spread of mechanically induced Ca²⁺ waves in unit G epithelial sheets at all (and only slightly inhibited Ca²⁺ waves of sheets kept in sµG). It seems possible that the epithelial sheets have decreased PKC to some extent; however, the cells still maintain the PKC necessary for the ATP responses. The cells of epithelial sheets still show ATP-induced Ca²⁺ oscillations, which most likely depend on PKC-negative feedback (29) and are sensitive to TPA inhibition. It is possible that the putative effect that TPA has to inhibit gap junctional-dependent Ca²⁺ waves in outgrowth cultures is overcome by a dominant extracellular communication of Ca²⁺ waves in epithelial sheets.

In tracheal epithelial outgrowth cultures, Ca²⁺ waves appear to spread primarily by diffusion of an intercellular messenger from the mechanically stimulated cell through gap junctions to neighboring cells (2, 16, 28). In astroglial cells, TPA impedes the spread of Ca²⁺ waves by causing a reduction in gap junctional permeability (8), and TPA may act similarly in tracheal epithelial outgrowth cultures (29). In contrast, mechanically stimulated Ca²⁺ waves appear to spread by release and diffusion of an extracellular messenger in hepatocyte (26), glial (4, 14), and mammary epithelial (9) cultures. In two experiments, Ca²⁺ wave communication between physically separated cells suggests that an extracellular messenger, instead of or in addition to an intracellular messenger, spreads Ca²⁺ waves in epithelial sheets. Extracellular propagation of Ca²⁺ waves in epithelial sheets would be insensitive to a putative TPA-dependent inhibition of gap junctional permeability. In the presence of continuous flow of sHBSS across epithelial sheets, the Ca²⁺ response to local application of ATP was biased in the direction of the flow, whereas mechanically stimulated Ca²⁺ wave propagation was not biased (see RESULTS). The absence of bias in Ca²⁺ waves under fluid flow is consistent with Ca²⁺ communication via an intercellular messenger in epithelial sheets as in outgrowth cultures; however, this is a negative result and more work is necessary to determine the mechanism of Ca²⁺ wave propagation in epithelial sheets. If the two cell preparations use different mechanisms, it will be important to determine which is closer to in situ biology.

In summary, sµG potentiates the inhibitory effects of TPA, possibly via an increase in PKC. Although others have made similar suggestions in different cell types, it remains unclear how microgravity alters PKC. If PKC is affected by mechanical stress and by the microgravity condition of spaceflight, it could serve as an assay for and as an explanation of some of the detrimental biological effects observed in astronauts.