Robert M. Davidson 1, Dimitris W. Tatakis 1, and Anthony L. Auerbach 2. Departments of i ... With regard to bone cells, Duncan and Misler (1989) have reported ...
PflfigersArch (1990) 416:646-651
Journal of Physiology 9 Springer-Verlag1990
Multiple forms of mechanosensitive ion channels in osteoblast-like cells Robert M. Davidson 1, Dimitris W. Tatakis 1, and Anthony L. Auerbach 2 Departments of i Oral Biologyand 2 BiophysicalSciences,Universityat Buffalo,Buffalo,NY 14214, USA Received September 26, 1989/Receivedafter revision February 19/AcceptedApril 23, 1990
Abstract. Patch-clamp recording techniques were used to examine the direct effects of mechanical stimulation on ion channel activity in human osteoblast-like osteosarcoma cells. Three classes of mechanosensitive ion channels were present and could be distinguished on the basis of conductance, ionic selectivity, and sensitivity to membrane tension. The largest conductance channel (160 pS) was K § and showed both a decrease in long closed interval duration and an increase in burst length with increasing membrane tension. For low applied pressures, there was an e-fold increase in the probability of this channel being open (Popen) for every 3.4 cm 2 Hg change in pressure. Two additional pressure-dependent channels had smaller conductances, i.e., 60 pS and 20 pS; the 60 pS channel appeared to be non-selective for cations. We propose that one or more of these mechanosensitive channels is involved in the response of bone to mechanical loading. Key words: Osteoblasts - Ion channels transduction - Stretch-activation
Mechano-
Introduction Mechanical loading of bone results in tissue remodelling (Lanyon 1984). In adult rat and avian bone in vivo, the application of mechanical stress increases the number of osteogenic layers and the width of the periosteum, and causes a proliferative response in osteoblasts leading to the formation of new bone (Pead et al. 1988; Miyawaki and Forbes 1987). In vitro, the application of mechanical stress to bone ceils grown in tissue culture leads to elevated levels of cyclic 3',5'adenosine monophosphate (cAMP) (Rodan and Rodan 1984; Rodan et al. 1975), increased thymidine incorporation (Shimshoni et al. 1984), and enhanced synthesis of prostaglandin E2 (Somjen et al. 1980; Yeh and Rodan 1984). While little is known about the mechanisms that link mechanical stimulation with either the tissue or cellular Offprint requests to: R.M. Davidson, Department of Periodontology,Universityof ConnecticutHealthCenter, Farmington, CT 06032, USA
response, evidence suggests that alterations in ion channel activity in the osteoblast membrane are associated with bone cell activation. Parathyroid hormone (PTH), a potent stimulator of bone resorption that alters osteoblast morphology (Galliard et al. 1979), collagen synthesis (Kream et al. 1980, 1986) and calcium metabolism (Yamaguchi et al. 1987), has been shown to either hyperpolarize (Chow et al. 1984; Ferrier and Ward 1986) or depolarize (Edelman et al. 1986; Fritsch et al. 1988) tissue-cultured osteoblasts, Other agents, such as cAMP, prostaglandin E2, hydrocortisone, and calcitonin also have been shown to alter membrane potential in these cells (Ferrier et al. 1985; Chow et al. 1984). Coupled with these observations are results showing that osteoblast membranes contain a wide variety of ion channels whose activity could underlie the modulation of the membrane potential. Osteoblasts have been shown to have a voltagegated, tetrodotoxin-sensitive Na + channel (ChesnoyMarchais and Fritsch 1988), several types of voltagegated Ca 2 -- channels (Guggino et al. 1988; Yamaguchi et al. 1987; Chesnoy-Marchais and Fritsch 1988; Ypey et al. 1988), and several types of K+-selective channels, including a Ca2+-activated K § channel (Dixon et al, 1984). Several of these channels have both transient and sustained voltage-controlled conductances (Ypey et al. 1988). In the present study, patch-clamp recording techniques were used to examine the electrophysiological response of osteoblast membranes to applied mechanical stress. Many cell types that might be expected to respond to mechanical deformation, including skeletal muscle (Guharay and Sachs 1984), smooth muscle (Kirber et al. 1988), cochlear hair cells (Ohmori 1984), and renal proximal tubule (Sackin 1989) contain mechanosensitive ion channels (see reviews by Sachs 1988; Morris 1989). With regard to bone cells, Duncan and Misler (1989) have reported the presence of a 20 pS, Ca z § mechanosensitive channel in the UMR-106 osteosarcoma cell line, and Yamaguchi et al. (1989) have shown that cell-swelling induces a Ca 2 § conductance in these cells. Our results indicate that human osteoblast-like cells contain several types of mechanosensitive channel that open in response to increased membrane tension. Some of these results have been presented in abstract form (Davidson et al. t989).
647
Methods
A
Human osteoblast osteosarcoma G292 (no. CRL 1423) cells were obtained from American Type Culture Collection (ATCC) (Rockville, Md.) and maintained at 37~ with McCoy's 5a supplemented with 10% fetal calf serum, penicillin, and streptomycin in a 5% COz humidified incubator. These cells were chosen because of their characterized osteoblast-like phenotype (Peebles et al. 1978 ; Shupnik and Tashjian 1982). Cells of various passages were seeded at 5 x 104 cells/ml (1 ml/dish; 35-ram dishes), and then incubated for 4 - 7 days. On the day of recording, the culture medium was gradually exchanged with standard extracellular saline solution [in mM: 150 NaC1, 3 KC1, 4 CaC12, 1 MgC12, 10 4-(2-hydroxyethyl)l-piperazine ethane sulfonic acid (HEPES) at 22-25~ Patchclamp recording pipettes were filled with either standard extracellular saline, a high KC1 solution [153 KCI, 2 MgClz, 1 CaC12, 11 ethylenebis(oxonitrilo)tetraacectate (EGTA), 10 HEPES/KOH, pCa7], or mixed KC1/NaC1 solution (78 NaC1, 75 KC1, 2 MgClz, I CaC12, 11 EGTA, 10 HEPES/KOH). For some experiments the culture medium was exchanged for Ca 2 +-free saline (150 NaC1, 3 KC1, 2 MgC12, 1 CaC12, 11 EGTA, 10 HEPES/KOH); during these experiments cells remained in Ca2+-free medium throughout the recording (typically 1 - 3 h). The pH of all solutions was adjusted to 7.4. All patches were in the cell-attached configuration (Hamil et al. 1981). Pipette resistances ranged from 4 to 12 MQ. Single-channel currents were monitored with a List EPC-7 patch-clamp amplifier (Darmstadt-Eberstadt, FRG), and stored on tape using a conventional VHS video recorder and a modified digital audio processor (Sony PCM-701ES, Tokyo, Japan). Once a giga-seal was formed, pressure was applied through the pipette using a 20 cm 3 syringe, and monitored with a mercury manometer in parallel with the pipette. Changes in pressure typically took 2--3 s to stabilize. The applied pressure is expressed as cm Hg (suction negative). For analysis, analog signals were filtered with an eight-pole lowpass Bessel filter (Frequency Devices, Haverhill, Mass.) at a cut-off frequency of 2 kHz or 5 kHz and digitized at a sampling rate of 10 kHz or 20 kHz. Data were collected for 1 - 3 min at each pressure. Single-channel currents were detected by automated analysis programs (IPROC and LPROC, Axon Instruments, Burlingame, Calif.). Data sets were fitted by appropriate functions with a nonlinear least-square algorithm (NFITS, C. Lingle, Washington Univ., St. Louis, Mo.). For interval duration histograms, the residuals were weighted by the inverse of the number of counts in each bin; for the rate constant vs pressure data (Fig. 4C) the residuals were weighted by the inverse of the 90% linear confidence limit on each rate constant estimate. For all other data sets, uniform weighting was used. Frequency histograms of current amplitude were fitted by sums of Gaussians, and mean current amplitudes and the probabilities of a single channel being open (Povo,) were derived from the fitted values (Yang 1989): Poper~ :
I --
Pclosed1/n,
where Pclosedis the probability of not being open in the record (equal to the number of samples in the baseline peak divided by the total samples in the record), and n is the number of channels in the patch, taken as the largest integral multiple of the unitary current level. For very small channel currents, discrete peaks in the amplitude histogram were not always apparent, and the current amplitude was estimated as the mean amplitude of single channel currents detected by visual inspection.
Results E i g h t y - f o u r p e r c e n t (34/40) o f the p a t c h e s s h o w e d evidence o f m e c h a n o s e n s i t i v e channels. I n the a b s e n c e o f a p p l i e d pressure, the a v e r a g e Pc~osed was 0.995 (range = 0 . 7 9 4 - 1.000; n = 22 patches). W h e n - 1 to - 4 c m H g was a p p l i e d to the p a t c h there was a m a r k e d increase in
25 ms
looooo
o cmHg Popen = 0 " 0 0 7
9 and the probability of being open within a burst (Pob, ,i) did not vary significantly with the applied pressure
three additional patches showed similar trends in the pressure dependencies of ~c3 and %2. There are severaI ways in which three closed and one open state can be connected in a Markovian kinetic model. However, the clear bursting activity of the channel suggests that these kinetic schemes can be reduced to a simple two-state system:
A second estimate of the steady-state pressure sensitivity was derived from the Pop~, values as determined from fitting amplitude histograms (e.g., Fig. 1 C). Data were fitted by a Boltzman equation with an offset to allow for inactivation: Pmax
PoP~" - 1 + Kexp (-- O~PZ), -- Pinaelive
//
{c,} v
,
/7 =/7o e x p ( - O~p2),
(I)
~z= eo exp(~ O~PZ),
(2)
K~q = (C~o/flo)exp[(- O~ + Op)P 2]
(3)
where C~ represents interburst closures, C2 and Ca intraburst closures, P is the applied pressure, O the sensitivity to pressure, and c% and ]70 are the transition rates in the absence of applied pressure. The data in Fig. 4C show that long closed intervals and long bursts had roughly equal but opposite sensitivites to pressure, with 8.5 cm 2 required to produce an e-fold decrease in the long closed interval duration (Off ~ 0A2 _4-0.03) and 6.9 cm 2 required to produce an e-fold increase in long burst durations (O~ = - 0 . 1 4 _+ 0.05). From these results we estimate the pressure sensitivity of the steady-state response, O~,, as the sum of ( - Ocz + Off), i.e. 0.26/era z.
where P~,x is the probability of being open within a burst (fixed at 0.86; see Fig. 4D) and Pin,~,ive is the probability that a channel is in an inactive state (assumed to be independent of membrane tension). The calculated fit is superimposed on the Pop~ data in Fig. 4A, with O~ equal to 0.29/cm 2 (3.4 cm 2 for an e-fold increase in Pop,n) and Pi~,~ti~ equal to 0.07. For a total of eight separate patches, O~ was 0.28 4- 0.28, and Pi,~,u~0 was 0.07 + 0.14.
Discussion The major finding in this study was the identification of three mechanosensitive cation channels in osteoblast-like cells showing both K +-selective (7t6o) and non-selective (760 and 72o) permeation properties. Although multiple forms ofmechanosensitive ion channel in single cells have been observed (Morris and Sigurdson 1989; Ding et al. i989), most cells with mechanosensitive currents appear
650 to have only a single class of channet (Morris 1989). The co-existence ofmechanosensifive non-selective cation and K+-selective channels in these cells suggests that a mechanical stimulus could induce either membrane hyperpolarization, depolarization, or a multiphasic response depending on the density and/or kinetic properties of the channels in tlhe deformed region of the membrane. Further, a non-uniform distribution of these channels along the axis of the cell could lead to a complex, spatially divergent response to whole-celt stretch. In physiological concentrations of calcium, membrane stretch would be expected to activate a Ca 2 + conductance in osteoblasts (Yamaguchi et al. 1989; Duncan and Misler 1989) that might activate the high-conductance Ca 2 +-dependent K + channel also present in these cells (Dixon et at. 1984; and see Ubt et al. 1988a; Christensen 1987; Taniguchi and Guggino 1989). In our experiments, all three forms ofmechanosensitive channel could be activated by stretch in the absence of extracellular Ca 2+, so we conclude that channel activation in the cell-attached configuration does not depend on influx of extracellular Ca 2+. We cannot, however, rule out the possibility that mechanical stimulation induces the release of calcium (or some other messenger) from intracellular stores (Snowdowne 1989), resulting in activation of one or more of these channels. Thus, we are unable to distinguish between a direct effect of membrane tension on channel gating (molecular mechanotransduction) and an indirect effect mediated by intracellular messenger(s) (cellular mechanotransduction). In particular, the similarity of the 7160 channels to Ca 2 +-activated K + channels that are present in osteoblasts (Ypey et al. 1988) offers the possibility that these two channels are the same. In either event, whether via direct or indirect mechanisms, ";~6o channels are likely to be activated by mechanical stimulation of osteoblasts in vivo, and would presumably modulate whole-cell currents produced by mechanical loading of these cells. Kinetic analysis of 716o channels revealed that activation and inactivation rates showed approximately equal and opposite sensitivities to applied pressure; the opposite polarity of the pressure-sensitivities of these rates suggests that the energy wells of both open (burst) and closed states are affected by mechanical stimulation. If ~t6o channels are indeed, directly activated by membrane stretch, the similarity in the magnitude of these sensitivities suggests that the compliances of the open and closed states are alike. This behavior is different from that of mechanosensifive ion channels in skeletal muscle (Guharay and Sachs 1984), proximat tubule (Sackin t989), and cardiac cells (Sigurdson et at. 1987), where long closed intervals decrease, but burst durations remain essentially unchanged (but see Kirber et al. 1987). Thus, for example, 7160 channels are approximately twice as sensitive as mechanosensitive channels seen in chick skeletal muscle, i,e., O,~ = 0.26/cm 2 in osteoblasts vs O~ = 0.14/cm 2 in chick skeletal muscle (Guharay and Sachs 1984). Moreover, neither chick nor toad muscle mechanosensitive channels show evidence of inactivation, whereas 7t6o channels were inactive about 7% of the time.
The whole-cell currents resulting from mechanical deformation depend on channel types, densities (n), kinetics (Pope,), and single-channel current amplitudes. With respect to the whole-cell current in the osteoblast, during mechanical stimulation the larger currents carried by ? t 6o and 760 channels may be offset by the apparent higher density and greater stretch sensitivity of the 720 currents. For the record shown in Fig. 1 (upper trace), Popon was 0.38 for the ?20 channel, versus 0.073 for the 7160 channel, approximately a 5-fold difference. In addition, there was often more 7'20 channels present than 716o and 7'6o channels. Thus, Y2ochannels may carry" a higher proportion of the stretch-activated current than its conductance alone would suggest, and these channels may make a major contribution to net ionic flux during mechanical loading. The timing of mechanosensitive ion channels in bone cells is particularly interesting since recent studies have shown (1)that mechanical loading is capable of activating osteoblasts (Pead et al. 1988) and (2) that ion channel activity is associated with early events in bone celt metabolism (Yamaguchi et at. 1987; Guggino et al. 1988; Edelman et at. 1986). Mechanosensitive ion channels have now been identified in cells from a variety of tissues (Sachs 1988; Morris t989), and stretch-activated channels may play an important role in such diverse functions as the transduction of auditory stimuli (Ohmori 1988), fluctuations of vascular pressure (Lansman et at. 1987), the contractility of smooth muscle fibers (Kirber et al. 1988), or ceU-volume regulation (Sackin 1989; Ubl et al. 1988b; Christensen 1987). Bone tissue responds directly to mechanical stress (Peak et al. 1988); shortterm mechanical loading of bone cells results in increased levels of cyclic nucleotides (Somjen et al. 1980; Shimshoni et al. 1984), increased production of prostaglandins (Somjen et al. 1980; Yeh and Rodan 1984), and morphological changes associated with remodelling of bone tissue (Pearl et at. 1988). While we cannot easily relate the stresses we have applied via suction to those that might be encountered in vivo (Sachs 1988), it is reasonable to assume that all three types of mechanosensitive channel would increase activity during mechanical loading of whole bone, in particular since osteoblast cells form a monolayer overlying calcified tissue (Marks and Popoff 1988; Raisz and Kream 1983). We propose that one or more of these mechanosensitive ion channels mediates the response of bone to mechanical stimulation, although the precise relationship between channel activation and bone remodelling remains unclear.
Acknowledgements. Tiffs work was supported by U.S.P.H.S. grants to R. M. D. (DE00145) and A. L~A. (NS2352301); D. N. T was supported by U. S. P. H. S. grant DE07034. We wish to thank Fred Sachs for helpful discussionsand Robert Borschetfor technical assistance. References
Chesnoy-Marchais D, Fritsch J (1988) Voltage-gated sodium and calcium currents in rat osteoblasts. J Physiol (Lond) 398:29/-311 Chow SY, Chow YC, Jee WSS, Woodbu~~ DM (t984) Electrophysiological properties of osteoblastlike cells from the
651 cortical endosteal surface of rabbit long bones. Calcif Tissue Int 36:401-408 Christensen O (1987) Mediation of cell volume regulation by Ca 2+ influx through stretch-activated channels. Nature 330: 6 6 - 69 Davidson RM, Tatakis DN, Dziak R, Auerbach A (1989) Stretchactivated ion channels in an osteoblast-like tumor cell line. J Bone Miner Res 4 [Suppl 1]: $329 Ding JP, Bowman CL, Sokabe M, Sachs F (1989) Mechanical transduction in glial cells: SACS and SICS. Biophys J 55:244a Dixon SJ, Aubin JE, Dainty J (1984) Electrophysiology of a clonal osteoblast-like cell line: evidence for the existence of a Ca 2+activated K + conductance. J Membr Biol 80:49-58 Duncan R, Misler S (1989) Voltage-activated and stretch-activated Ba2§ conducting channels in an osteoblast-like cell line (UMR 106). FEBS Lett 251:17-21 Edelman A, Fritsch J, Balsan S (1986) Short-term effects of PTH on cultured rat osteobla sts: changes in membrane potential. Am J Physiol 251:C483-490 Ferrier J, Ward A (1986) Electrophysiological differences between bone cell clones: membrane potential responses to parathyroid hormone and correlation with the cAMP response. J Cell Physiol 126:237- 242 Ferrier J, Illeman A, Zakshek E (1985) Transient and sustained effects of hormones and calcium on membrane potential in a bone cell clone. J Cell Physiol 122:53 - 58 Fritsch J, Edelman A, Balsan S (1988) Early effects of parathyroid hormone on membrane potential of rat osteoblasts in culture: role of cAMP and Ca z § J Bone Miner Res 3:547-554 Gaillard PJ, Hermann-Erlee MPM, Hekkelman JW, Burger EH, Nijweide PJ (1979) Skeletal tissue in culture: hormonal regulation of metabolism and development. Clin Orthop 142:176214 Guggino SE, Wagner JA, Snowman AM, Hester LD, Sacktor B, Snyder SH (1988) Phenylalkylamine-sensitivecalcium channels in osteoblast-like osteosarcoma cells. J Biol Chem 263:10155 10161 Guharay F, Sachs F (1984) Stretch-activated single ion channel currents in tissue-cultured embryonic chick skeletal muscle. J Physiol (Lond) 352: 6 8 5 - 701 Hamill OP, Marty A, Neher E, Sakmann B, Sigworth F (1981) Improved patch-clamp techniques for high resolution current recording from cells and cell-free membrane patches. Pfl/igers Arch 391:85-100 Kirber MT, Walsh JV, Singer JJ (1988) Stretch-activated ion channels in smooth muscle: a mechanism for the initiation of stretchinduced contraction. Pfl/igers Arch 412: 339- 345 Kream BE, Rowe D, Gworek SC, Raisz L (1980) Parathyroid alters collagen synthesis and procollagen RNA levels in fetal rat calvaria. Proc Natl Acad Sci USA 77:5654-5658 Kream BE, Rowe D, Smith MD, Majeska R (1986) Hormonal regulation of a collagen synthesis in a rat clonal osteosarcoma cell line. Endocrinology 199:1922-1928 Lansman JB, Hallam J, Rink TJ (1987) Single stretch-activated ion channels in vascular endothelial cells as mechanotransducers? Nature 327 : 524 - 526 Lanyon LE (1984) Functional strain as a determinant for bone remodelling. Calcif Tissue Int 36: $ 5 6 - $61 Marks SC, Popoff SN (1988) Bone cell biology: the regulation of development, structure, and function in the skeleton. Am J Anat 183: i - 4 4 Miyawaki S, Forbes DP (1987) The morphologic and biochemical effects of tensile force application to the interparietal suture of the Sprague-Dawley rat. Am J Orthod Dentofacial Orthop 92:123-133 Morris CE (1989) Mechanosensitive ion channels. J Membr Biol (in press) Morris CE, Sigurdson WJ (1989) Stretch-inactivated ion channels coexist with stretch-activated ion channels. Science 243:807809
Ohmori H (1984) Mechanoelectrical transducer has discrete conductances in the chick vestibular hair cell. Proc Natl Acad Sci USA 81 : 1888-1891 Pead MJ, Skerry TM, Lanyon LE (1988) Direct transformation from quiescence to bone formation in the adult periosteum following a single brief period of bone loading. J Bone Miner Res 3:647--656 Peebles PT, Trisch T, Papageorge AG (1978) Isolation of four unusual pediatric solid tumor cell lines. Pediatr Res 12:485 Raisz LG, Kream BE (1983) Regulation of bone formation. N Eng J Med 309:29-35 Rodan GA, Rodan SB (1984) Expression of the osteoblastic phenotype. In: Peck, WA (ed) Bone and mineral research, vol 2. Excerpta Medica, Amsterdam, pp 244--285 Rodan GA, Bourret LA, Harvey A, Mensi T (1985) Cyclic AMP and cyclic GMP: mediators of the mechanical effects of bone remodelling. Science 189 :4 6 7 - 469 Sachs F (1988) Mechanical transduction in biological systems. CRC Crit Rev Biomed Eng 16:141-169 Sackin H (1989) A stretch-activated K + channel sensitive to cell volume. Proc Natl Acad Sci USA 86:1731-1735 Shimshoni Z, Binderman I, Fine N, Somjen D (1984) Mechanical and hormonal stimulation of cell cultures derived from young rat mandible condyle. Arch Oral Biol 29:827-837 Shupnik MA, Tashjian AH Jr (1982) Epidermal growth factors and phorbol ester actions on human osteosarcoma cells: characterization of responsive and nonresponsive cell lines. J Biol Chem 257:1216-1264 Sigurdson WJ, Morris CE, Brezden BL, Gardner DR (1987) Stretch activation of a K § channel in molluscan heart cells. J Exp Biol 127:191 --209 Sigworth FJ, Sine SM (1987) Data transformations for improved display and fitting of single-channeldwell time histograms. Biophys J 52:1047-1054 Snowdowne KW (1987) The effects of stretch on sarcoplasmic free calcium of frog skeletal muscle at rest. Biochim Biophys Acta 862: 441 - 444 Somjen D, Binderman I, Berger E, Harell A (1980) Bone remodelling induced by physical stress is prostaglandin E2 mediated. Biochim Biophys Acta 627:91 -100 Taniguchi J, Guggino WB (1989) Membrane stretch: a physiological stimulator of Ca2+-activated K § channels in thick ascending limb. Am J Physiol 257: F 3 4 7 - F352 Ubl J, Murer H, Kolb H-A (1988a) Hypotonic shock evokes opening of Ca2+-activated K channels in opossum kidney cells. Pfliigers Arch 412:551 - 553 Ubl J, Murer H, Kolb H-A (1988b) Ion channels activated by osmotic and mechanical stress in membranes of opossum kidney cells. J Membr Biol 104:223-232 Yamaguchi DT, Hahn T J, Iida-Klein A, Kleeman CR, Muallem S (1987) Parathyroid hormone-activated calcium channels in an osteobtast-like clonal osteosarcoma cell line. J BioI Chem 262:7711-7718 Yamaguchi DT, Green J, Kleeman CR, Muallem S (1989) Characterization of volume-sensitive, calcium-permeating pathways in the osteosarcoma cell line UMR-106-01. J Biol Chem 264:4383 -4390 Yang X (1989) Characterization of stretch-activated ion channels in Xenopus oocytes. Ph. D. thesis, University at Buffalo Yeh C, Rodan GA (1984) Tensile forces enhance prostaglandin E synthesis in osteoblastic cells grown on ribbons. Calcif Tissue Int 36 : $ 6 7 - $71 Ypey DL, Ravesloot JH, Buisman HP, Nijweide PJ (1988) Voltageactivated ionic channels and conductances in embryonic chick osteoblast cultures. J Membr Biol 101 : 141 - 150
