Characterization of a stretch-activated potassium channel in chondrocytes

Abstract

Chondrocytes possess the capacity to transduce load-induced mechanical stimuli into electrochemical signals. The aim of this study was to functionally characterize an ion channel activated in response to membrane stretch in isolated primary equine chondrocytes. We used patch-clamp electrophysiology to functionally characterize this channel and immunohistochemistry to examine its distribution in articular cartilage. In cell-attached patch experiments, the application of negative pressures to the patch pipette (in the range of 20-200 mmHg) activated ion channel currents in six of seven patches. The mean activated current was 45.9 +/- 1.1 pA (n = 4) at a membrane potential of 33 mV (cell surface area approximately 240 microm(2)). The mean slope conductance of the principal single channels resolved within the total stretch-activated current was 118 +/- 19 pS (n = 6), and reversed near the theoretical potassium equilibrium potential, E(K+), suggesting it was a high-conductance potassium channel. Activation of these high-conductance potassium channels was inhibited by extracellular TEA (K(d) approx. 900 microM) and iberiotoxin (K(d) approx. 40 nM). This suggests that the current was largely carried by BK-like potassium (MaxiK) channels. To further characterize these BK-like channels, we used inside-out patches of chondrocyte membrane: we found these channels to be activated by elevation in bath calcium concentration. Immunohistochemical staining of equine cartilage samples with polyclonal antibodies to the alpha1- and beta1-subunits of the BK channel revealed positive immunoreactivity for both subunits in superficial zone chondrocytes. These experiments support the hypothesis that functional BK channels are present in chondrocytes and may be involved in mechanotransduction and chemotransduction.

Fig. 1

Activation of ion channels by membrane stretch. A: Application of suction (as indicated by the solid bar) to the patch pipette of chondrocytes under cell-attached mode of recording increases membrane current. V_m = 33 mV. B: Similar experiment to “A,” but with 3 mM TEA included in the patch pipette. The suction “response” is greatly reduced. C: Increasing negative pressures yield increasingly large activations of current. The y-axis is the percentage current activation of the maximum seen in each particular patch. Data from nine patches. D: Single channels (V_m = 33 mV) in the “tail” of the suction response for a control experiment similar to that in “A.” E: Single channels (V_m = 33 mV) in the “tail” of the suction response for an experiment similar to that in “B,” with 3 mM TEA in the patch pipette. F: Summary of experiments similar to that shown in A and B. *P ≤ 0.05. IBTX, iberiotoxin.

Fig. 2

Stretch activates a high-conductance potassium current. A–F: cell-attached patch recordings of stretch-activated channel activity with 115 mM pipette [K⁺]. A,C,E: Raw traces at −40, +20, and +40 mV respectively. B,D,F: All-points amplitude histograms for the events shown in corresponding traces A, C, and E. Results from similar experiments are shown in G to J, but with 15 mM [K⁺] in the pipette. K: Mean current-voltage curves from experiments similar to those illustrated in A to J. Triangles: pipette 115 mM (n = 6), circles: pipette 15 mM (n = 5). The straight line fitted through the 115 mM K⁺ data has slope 192 pS. The 15 mM K⁺ data are fit with the Hodgkin–Katz current equation (Hodgkin and Katz, 1949) assuming the presence of only potassium conductance with P_K+ 0.4e⁻¹² m³ s⁻¹. Complete solutions and calculated E_K+ values are described in Table 1.

Fig. 3

Isolated patch recordings. A: Application of negative pressure to the side port of the pipette holder stretches the membrane beneath the patch pipette (cell-attached patch mode) and activates an ion channel current (−3 mV). B: Channel activity is still apparent following patch excision (inside-out patch) and is clearly sensitive to 200 µM cytoplasmic calcium or 500 µM EGTA (−3 mV). C: Ion channel activity activated as described in A and B (inside-out patch, 200 µM Ca²⁺, but with 145/55 mM K⁺ E_K+ = 25 mV) at a range of holding potentials. D: Current-voltage curves from nine experiments such as that shown in C, in inside-out patch mode. The curve represents a straight line regression with slope 147 pS.

Fig. 4

Whole-cell stretch hyperpolarizes chondrocytes. A: Application of membrane stretch by means of hypotonic challenge (from 314 to 134 mOsm) hyperpolarizes the membrane. Membrane potential was measured continuously with periodic injections of current to monitor cell integrity. B: An equivalent experiment repeated (different cell) in the presence of 10 mM TEA. The hyperpolarization is significantly reduced by the presence of 10 mM TEA (B,C). *P ≤ 0.05.

Fig. 5

Distribution of the α1- and β1-subunits of the BK channel (KCNMB1 and KCNMNB1) in equine articular cartilage. Immunohistochemical analysis of samples of full-depth equine articular cartilage was carried out using polyclonal antibodies raised against the α1- and β1-subunits of the BK channel. Sections of equine cartilage were immunostained with primary antibodies and horseradish peroxidase-labelled rabbit anti-goat secondary IgG (DakoCytomation). Positive immunoreactivity for both subunits was predominantly observed in superficial zone chondrocytes in normal cartilage. The magnified areas in the insets shown in panels A and B highlight the chondrocyte-specific immunostaining. Omission of primary antibody from the immunohistochemical procedure served as negative controls. Sections of equine articular cartilage were treated in exactly the same way during the immunohistochemical procedure except that the primary antibody was omitted. Original magnifications of the main panels: 200×. Bars in the main panels represent 100 µm. Bars in the magnified insets shown in panels A and B represent 10 µm.