- Email: [email protected]
Influence of external pH on two types of low-voltage-activated calcium currents in primary sensory neurons of rats
Biochimica et Biophysica Acta 1724 (2005) 1 – 7 http://www.elsevier.com/locate/bba
Influence of external pH on two types of low-voltage-activated calcium currents in primary sensory neurons of rats V.O. Pinchenko*, P.G. Kostyuk, E.P. Kostyuk Department of General Physiology of Nervous System, Bogomoletz Institute of Physiology, The National Academy of Science of Ukraine, Bogomoletz Street 4, Kyiv 01024, Ukraine Received 3 March 2004; received in revised form 29 March 2005; accepted 5 April 2005 Available online 25 April 2005
Abstract The influence of extracellular pH (pHo) on low-voltage-activated calcium channels of acutely isolated DRG neurons of rats was examined using the whole cell patch-clamp technique. It has been found that in the neurons of middle size with capacitance C = 60 T 4.8 pF (mean T S.E., n = 8) extracellular acidification from pHo 7.35 to pHo 6.0 significantly and reversibly decreased LVA calcium current densities by 75 T 3.7%, shifted potential for half-maximal activation to more positive voltages by 18.7 T 0.6 mV with significant reduction of its voltage dependence. The half-maximal potential of steady-state inactivation shifted to more positive voltages by 12.1 T1.7 mV (n = 8) and also became less voltage dependent. Dose – response curves for the dependence of maximum values of LVA currents on external pH in neurons of middle size have midpoint pK a = 6.6 T 0.02 and hill coefficient h = 0.94 T 0.04 (n = 5). In small cells with capacitance C = 26 T 3.6 pF (n = 5), acidosis decreased LVA calcium current densities only by 15.3 T 1.3% and shifted potential for half-maximal activation by 5.5 T 1.0 mV with reduction of its voltage dependence. Half-maximal potential of steady-state inactivation shifted to more positive voltages by 10 T 1.6 mV (n = 4) and also became less voltage dependent. Dose – response curves for the dependence of maximum values of LVA currents on external pH in neurons of small size have midpoint pK a = 7.9 T 0.04 and hill coefficient h = 0.25 T 0.1 (n = 4). These two identified types of LVA currents besides different pH sensitivity demonstrated different kinetic properties. The deactivation of LVA currents with weak pH sensitivity after switching off depolarization to 30 mV had substantially longer decay time than do currents with strong pH sensitivity (s d ¨5 ms vs. 2 ms respectively). It was found that the prolongation of depolarization steps slows the subsequent deactivation of T-type currents in small DRG neurons. Deactivation traces in these neurons were better described by the sum of two exponentials. Thus, we suppose that T-type channels in small DRG neurons are presented mostly by a1I subunit. We suggest that these two types of LVA calcium channels with different sensitivity to external pH can be differently involved in the origin of neuropathic changes. D 2005 Elsevier B.V. All rights reserved. Keywords: Dorsal root ganglion; T-type Ca2+ channels; Acidosis; pH sensitivity; Neuropathy
1. Introduction Diabetic neuropathy is a condition in which hyperglycemia is combined with hypoxia. Both of these factors can induce intracellular and/or extracellular acidification [1]. These pathophysiological changes may contribute to chronic neuropathic pain originating in nociceptors that are subpopulations of the primary sensory neurons (DRG neurons). * Corresponding author. Tel.: +7 38044 2562086; fax: +7 38044 2562093. E-mail address: [email protected] (V.O. Pinchenko). 0304-4165/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.bbagen.2005.04.008
It is well known that the DRG neurons can be divided into three groups according to their somatic sizes [2]. Nociceptors are commonly found among neurons of small and middle sizes and these neurons express both low-voltage-activated (LVA) and high-voltage-activated (HVA) calcium currents [2]. It has been shown [3] that extracellular acidification substantially attenuates inward calcium currents. In contrast to HVA calcium channel studies, only scant information exists about the regulation of LVA calcium channels. The pH sensitivity of native T-type currents has been previously studied only in few cases: in ventrobasal thalamic neurons [3], CA1 hippocampal neurons [4] and cardiac myocytes [5].
2
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
LVA Ca2+ channels start to open when cells are depolarized near 60 mV. They open and inactivate rapidly but their most remarkable kinetic property is that they close (deactivate) much slower than do HVA Ca2+ channels. Earlier, it has been shown [6] that in thalamic neurons, at least two types of LVA calcium channels with different kinetic and pharmacological profiles are expressed. Later, they were classified in 3 subtypes—Cav3.1, 3.2, and 3.3 [7]. Previously, a detailed analysis of the effect of pH changes was made on recombinant a1H [8] and a1G [9] T-type channels. Similar information on pH modulation of a1I subunit is still lacking. Such analysis on native T-type channels might be quite important because of the different location and functional significance of these subtypes. Therefore, we compared such effect on kinetically different T-types channels—with slow and fast inactivation and deactivation.
2. Methods 2.1. Cell preparation In our experiments, we used male 3-month-old Wistar rats. Neurons were freshly isolated from the lumbar dorsal root ganglia (DRG). Rats were decapitated after ether anesthesia, and the ganglia were removed and placed in Tyrode solution. After cleaning, they were incubated for 25 –30 min at 35- in the same solution supplemented with 1 mg/ml protease (Sigma) and 0.5 mg/ml collagenase Type 1 (Worthington). After, the enzymatic treatment ganglia were washed in the enzyme-free solution and dissociated by gentle pipetting. The cell suspension was plated on sterile glass coverslips and incubated in Tyrode solution for 60 min. It was enough to adhere a sufficient number of single cells to the coverslips to perform the recordings. The cells in the recording chamber were placed on the table of Zeiss inverted microscope with phase contrast objective and CCD camera connected to a TV monitor. The cell diameter was taken as a mean of longest and shortest diameters of cell body measured with a ruler on the monitor image. 2.2. Chamber and solutions The experimental chamber of 2 ml volume was continuously perfused (2 ml/min) with external bath solution composed of (mM): 2 CaCl2, 67 choline chloride, 100 tetraethylammonium chloride, 5.6 glucose, 5.3 KCl, 10 Hepes and 3.8 MgCl2. pH was adjusted to the desired value with KOH or HCl. The patch pipettes were filled with an electrode solution composed of (mM): 140 CsCl, 10 Hepes, 10 EGTA, 4 MgATP and 0.1 NaGTP (pH 7.2– 7.3 with CsOH).
Seal resistances >1 GV were achieved using patch pipettes of 1 –2 MV resistances when they were filled with an electrode solution. In all experiments, compensation of series resistance was performed using the standard procedure of the 3900A amplifier (Dagan Corp., Minneapolis, MN, USA). Data were acquired with DIGIDATA 1200 ADAC and amplified with 3900A amplifier using PCLAMP 6.0.3 (Axon Instruments Inc., USA) software, and final data records were analyzed with CLAMPFIT 8.0 program (Axon Instruments Inc., USA) and Microsoft Excel. Current signals were sampled at 20 kHz and filtered at 5 kHz through the interface connected to an AT-compatible computer system which also served as a stimulus generator. 2.4. Voltage protocols and data analysis Activation gating was measured by holding at 80 mV, then applying step to 90 mV for off-line leakage correction followed by V test from 60 mV to +50 mV. The peak current is plotted as a function of V test. We fit the current voltage curves with the Boltzmann equation modified with Golgman– Hodgkin – Katz equation: 0 1 B I Ca ðV Þ ¼ V B @
C G G 1 þ 2 C V1 V V2 V A 1 þ exp 1 þ exp k1 k2 2 3 zF ðV VCa Þ exp 1 6 7 RT 7 6 4 5 zFV 1 exp RT 2 3 zF ðV VL Þ exp 1 6 7 RT 7 ð1Þ þ V G3 6 4 5 zFV 1 exp RT
where G 1, G 2 and G 3 are maximal conductance for LVA, HVA and leakage currents, V 1, V 2, k 1, and k 2 are the midpoints and slope factors for LVA and HVA current activations, V Ca and V L are reversal potentials for calcium and leakage current components, T is absolute temperature, R is universal gas constant, F is the Faraday constant and z is the valence of a permeable ion (equal to 2 for Ca2+ and 1 for leakage currents). Steady-state inactivation is measured by holding at 80 mV, then pre-pulsing from 120 to 10 mV for 2.5 s, followed by a V test, to 30 mV. The peak current of V test is plotted as a function of pre-pulse potential. The data were fit with the Boltzmann equation:
2.3. Electrical recording
I 1 þC ¼ V V1=2 Imax 1 þ exp k
Voltage-clamp recordings using the whole-cell variant of the patch-clamp technique were made at room temperature.
where I/I max is relative current, V is pre-pulse potential, V 1/2 is midpoint potential for complete inactivation, k is the slope
ð2Þ
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
3
of the voltage dependence for inactivation, and C is a constant. The time constants of activation (s m) and inactivation (s h) of the currents were determined from a fit of current traces with the equation: I ðt Þ ¼ I o I½1 expð t=sm Þ2 I expð t=s h Þ þ I c
ð3Þ
where I o is a normalization factor and I c is a constant component. Deactivation time constants (s d) were determined from single-exponential fit of tail currents. Dose – response curves for the pH dependence of LVA currents were built using the following equation: I¼
Imax 1 þ 10ð pK a pH o Þh
ð4Þ
where I max is the maximal asymptotic value of T-current, pK a is pH value at half-maximal current and h is the Hill coefficient. Data are reported as mean T S.E. Student’s t tests on independent groups were used to evaluate P values.
3. Results DRG neurons of small and middle size possess both LVA and HVA calcium channels, with preferential expression of LVA channels in neurons of middle size. We did not use pharmacological tools to separate calcium currents because of the possible pH-dependent effects of the drugs. Thus, we made mathematical separation of LVA and HVA currents using Eq. (1) (see Methods). The example of such mathematical approach is shown in Fig. 1, where the I –V curve for total calcium current in representative cell of middle size was separated into LVA and HVA calcium current components together with leakage current for monovalent cations. Fig. 2 shows the mean I – V curves recorded in cells of middle (A) and small (B) size in control solution with pH 7.35 (filled squares) and under acidosis pH 6.0 (open circles). A striking difference in pH dependence of LVA components for these two types of cells can be clearly seen. The peak of the LVA component in middle size cells (A) was shifted by ¨20 mV in depolarized direction and decreased 5fold under acidosis, while in small cells (B), it decreased only to ¨50% of the initial value and shifted by ¨5 mV rightwards. To estimate the activation parameter changes under acidosis, we recorded I – V curves using 50 ms depolarizing steps with 5 mV increment from V hold = 80 mV. After the mathematical separation of LVA current from HVA current (Eq. (1)), we grouped the cells by the degree of acidosis influence on LVA component and named them as ‘‘strong’’ for those cells in which activation parameters changed substantially comparing with ‘‘weak’’ cells. It turned out that ‘‘strong’’ cells have larger somatic size with mean diameter and membrane capacitance 33 T 1 Am and 60 T 4.8 pF (n = 8), respectively, compared to
Fig. 1. The mathematical separation of experimental I – V curve (open circles) for total calcium current in a representative cell of middle size into LVA current (thin line), HVA current (dotted line) and leakage current (dashed line) using Eq. (1) (see Methods). The sum of these components (bold line) gives the best fit of the experimental I – V curve.
28 T 2 Am and 26 T 3.6 pF (n = 5) for ‘‘weak’’ cells. The influence of acidosis on activation parameters in these two groups of cells and the results of statistical analysis are presented on three panels of Fig. 3. As it can be seen in panel A, the activation midpoint potential was shifted under acidosis by ¨20 mV in the depolarized direction in ‘‘strong’’ cells and only by ¨5 mV in ‘‘weak’’ cells; on panel B, the activation slope factor was also changed to a larger extent in ‘‘strong’’ cells under acidosis (from 3.8 mV/e-fold to 5.2 mV/e-fold) and only from 5.8 mV/efold to 6.2 mV/e-fold in ‘‘weak’’ cells. The most striking difference in pH dependence of these two types of LVA currents on external pH is depicted on Fig. 4. To evaluate the pH dependence of T-currents in cells of middle and small size diameters, we have built the current –voltage curves for calcium currents in following external pHo values: 6.0, 6.5, 7.35, 8.0 and 8.5. Then, the maximum of T-currents at given pH was normalized to that at pH 7.35. Hill curves were plotted through the mean values obtained for several cells grouped by cell size (membrane capacitance). The pK a value and the Hill coefficient h were equal to 6.6 T 0.02 and 0.94 T 0.04 (n = 5) for middle size cells and to 7.9 T 0.04 and 0.25 T 0.1 (n = 4) for small size cells respectively. Thus, T-channels of the middle size cells are strongly influenced by external pH near normal pH 7.35. As the Hill coefficient is close to one and the value of pK a is shifted to the acidic pH value, the strong inhibition of T-currents took place under acidosis. In contrast, T-channels of small size cells exhibit low values of the Hill coefficient and the pK a value that is shifted to alkaline pH, which lead to relatively weak inhibition of the current through these channels under acidosis. Fig. 5 demonstrates the influence of external acidification on steady-state inactivation curves recorded in cells with strong pH sensitivity of LVA currents. External acidification to pHo 6.0 decreases the peak of LVA current, shifts the
4
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
Fig. 2. Mean I – V curves for calcium currents recorded in cells of middle size (n = 5; A) and small size (n = 4; B) in control solution with pH 7.35 (filled squares) and under acidosis pH 6.0 (open circles).
voltage dependence of steady-state inactivation and decreases its slope factor. Panels A and B show the family of currents in representative cell of middle size for pHo 7.35 and 6.0 elicited by a V test of 30 mV, following pre-pulses from 110 to 40 mV for 2.5 s. In panel C, peak currents from 8 middle-sized neurons are normalized to the peak of the maximally available current and relative current is plotted as a function of pre-pulse potential and is fitted with the Boltzmann distribution (solid line). The inactivation midpoint potential V mid is shifted from 69 mV at pH0 7.35 to 57 mV at pHo 6.0, with significant effect on the slope (5.4 mV/e-fold at pHo 7.35 and 7.6 mV/e-fold at pHo 6.0) (n = 8; P < 0.01). Fig. 6 shows the influence of acidification on steady-state inactivation curves recorded in cells with weak pH sensitivity of LVA currents. External acidification to pHo 6.0 decreases the peak of LVA current, shifts the voltage dependence of steady-state inactivation and slightly decreases its slope factor. Panels A and B represents a family of currents in typical cell of small size for pHo 7.35 and 6.0 elicited by the same protocol as in Fig. 5. Panel C
Fig. 3. Bar histograms summarizing acidosis influence on activation parameters in ‘‘strong’’ and ‘‘weak’’ pH-sensitive cells: the activation midpoint potential V mid in ‘‘strong’’ cells is equal to 50 T 1.0 mV in control and 31 T1.2 mV under acidosis; in ‘‘weak’’ cells, this value is equal to 48 T 0.8 mV in control and 43 T 0.7 mV under acidosis (A). The activation slope factor k in ‘‘strong’’ cells is equal to 3.8 T 0.3 mV/e-fold in control and 5.2 T 0.4 mV/e-fold under acidosis; in ‘‘weak’’ cells, this value is equal to 5.8 T 0.1 mV/e-fold in control and 6.2 T 0.1 mV/e-fold under acidosis (B).
shows the mean steady-state inactivation curves for 4 small neurons fitted with a Boltzmann distribution. The inactivation midpoint potential V mid is shifted from 80.5 mV at pHo 7.35 to 70.5 mV at pHo 6.0; also acidosis affects the slope factor, but to less extent than in cells with strong pH
Fig. 4. Dose – response of LVA currents on external pH in middle size cells (open circles) and in small size cells (filled squares). The best fit values of pK a and the Hill coefficient values according to Eq. (4) are shown on a legend.
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
5
Fig. 7. Kinetic properties of two representative traces of strong (bold line) and weak (dashed line) pH-sensitive LVA currents. Fig. 5. The influence of acidification on steady-state inactivation curves recorded in cells with strong pH sensitivity of LVA currents.
sensitivity (4.6 mV/e-fold at pHo 7.35 and 5.0 mV/e-fold at pHo 6.0). The kinetic properties of two representative traces of LVA currents recorded from strong and weak pHsensitive cells are shown in Fig. 7. Holding and test potentials were 80 mV and 30 mV, respectively. A dramatic difference in kinetic of inactivation and deactivation is obvious in these two cells, while the activation times remained almost the same. The voltage and pH dependence of activation and inactivation kinetic parameters for cells with strong and weak pH-sensitive LVA currents are shown in Fig. 8. The time constant for activation s m (panel A) in strong pH-sensitive cells was shifted by ¨20 mV in the direction of depolarization under acidosis, became less voltage dependent and increased approximately 3-fold near
the peak of the I – V curve (at 30 mV). In contrast to this observation, in cells with weak pH sensitivity (panel B), this shift reached only about ¨7 mV under acidosis, which is consistent with charge screening hypothesis. As for inactivation time constants (s h), similar conclusions could be done from the analysis of their pH and voltage dependence in cells with strong and weak pH sensitivity (panels C and D, respectively). In the paper [12], authors investigated a1I subunit expressed in HEK293 cells. They observed that the rate of deactivation at 80 mV was clearly dependent on the level of inactivation induced immediately prior to repolarization, with greater inactivation producing significantly slower deactivation. In our work, we used a similar protocol to investigate the possible influence of the extent of inactivation on the rate of deactivation of T-type currents in small DRG cells. An example of such protocol with current traces is shown on Fig. 9A. Test pulses to 45 mV of 18, 36, 70, 140 and 280 ms duration were applied from holding potential of 80 mV. The duration of repolarization to 80 mV was equal to 20 ms after every test pulse. We found that tail currents in small cells cannot be fitted well with single exponential, thus, we use for this purpose two exponentials. Fig. 9B shows the relationship of weighted average for fast and slow components of deactivation current on the duration of the prior test pulse to 45 mV. Small cells have T-type channels with properties similar to the above-mentioned cloned a1I subunits (strong correlation of deactivation s with test pulse duration).
4. Discussion
Fig. 6. The influence of acidification on steady-state inactivation curves recorded in cells with weak pH sensitivity of LVA currents.
In the present work, we have shown that primary sensory neurons possess at least two distinct types of LVA Ca2+ channels. These channels differ in pH sensitivity and
6
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
Fig. 8. Voltage and pH dependence of activation kinetic parameter (s m) for cells with strong (n = 5; A) and weak (n = 4; B) pH-sensitive LVA currents and inactivation kinetic parameter (s h) for cells with strong (C) and weak (D) pH-sensitive LVA currents at pH 7.35 (filled squares) and pH 6.0 (open circles).
Fig. 9. (A) The example of protocol and current traces used to find the relationship (B) between test-pulse duration and tail current kinetics.
kinetic properties. We have found that small DRG neurons express preferentially LVA Ca2+ channels with weak pH sensitivity while neurons of medium size express LVA Ca2+ channels with strong pH sensitivity. Also, weak pHsensitive channels have slower inactivation and deactivation kinetics than do strong pH-sensitive ones. In order to elucidate the possible a1 subunit composition of these channels, we compared our data with those obtained by other groups on recombinant Ca2+ channels (Cav3.2 [8] and Cav3.1 [9]). It was found that native strong pHsensitive LVA channels are similar to recombinant ones in terms of kinetic properties and influence of external acidification on midpoint potential of activation, but in case of the native channels, a more profound effect on the latter parameter can be seen. To our knowledge, there are no publications concerning the influence of external pH on the Cav3.3 subunit of LVA channels. The inactivation kinetics of weak pH-sensitive LVA Ca2+ channels in our study is similar to that of recombinant Cav3.3 channels [10], but the deactivation kinetics is much slower. Our experimental data indicate that the prolongation of depolarization steps slows subsequent deactivation. This is in good agreement with results obtained by Warre and Randall [12] on a1I subunit. Also, we have found that deactivation traces were better described by the sum of two exponentials. Earlier, Frazier et al. [13] concluded that a single exponential provides an adequate description of tail currents for a1G but not a1I subunits. Thus, we can conclude that the a1I subunit is mostly expressed in the case of small DRG neurons. Taking into account that DRG neurons express mostly Cav3.2 and Cav3.3 mRNA [7], we can suppose that the most probable
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
a1 subunit candidates for strong pH-sensitive LVA Ca2+ channels are the Cav3.2, while for weak pH-sensitive LVA Ca2+ channels, the Cav3.3. It has been recently shown that Cav3.2 channels can be also distinguished from other T-type channel subtypes by specific intracellular modulation by protein-kinase C [11], making them very sensitive to intracellular metabolic changes. The possible physiological significance of our findings can be considered in the terms of nociception. The DRG neurons of different size are known to possess different perception modality. The small neurons receive and transmit preferentially nociceptive signals, while neurons of middle and large sizes deal with proprioceptive sensations. Under pathological conditions of acidosis, the different pH sensitivity of LVA channels expressed in these neurons could lead to the shift of total signal modality to the nociceptive side, giving rise to different types of neuropathic changes. Acknowledgement This work was supported by INTAS grant No. 99-01915. References [1] K. Kaila, B.R. Ransom, pH and Brain Function, Wiley-Liss, Inc., 1998. [2] R.S. Scroggs, A.P. Fox, Calcium current variation between acutely isolated adult rat dorsal root ganglion neurons of different size, J. Physiol. (Lond.) 445 (1992) 639 – 658.
7
[3] M.J. Shah, S. Meis, T. Munsch, H.C. Pape, Modulation by extracellular pH of low- and high-voltage-activated calcium currents of rat thalamic relay neurons, J. Neurophysiol. 85 (2001) 1051 – 1058. [4] G.C. Tombaugh, G.G. Somjen, Differential sensitivity to intracellular pH among high- and low-threshold Ca2+ currents in isolated rat CA1 neurons, J. Neurophysiol. 77 (1997) 639 – 653. [5] J. Tytgat, B. Nilius, E. Carmeliet, Modulation of the T-type cardiac Ca2+ channel by changes in proton concentration, J. Gen. Physiol. 96 (1990) 973 – 990. [6] A.N. Tarasenko, P.G. Kostyuk, A.V. Eremin, D.S. Isaev, Two types of low-voltage-activated Ca2+ channels in neurons of rat laterodorsal thalamic nucleus, J. Physiol. (Lond.) 499 (1997) 77 – 86. [7] E.M. Talley, L.L. Cribbs, J.H. Lee, A. Daud, E. Perez-Reyez, D.A. Bayliss, Differential distribution of three members of gene family encoding low voltage-activated (T-type) calcium channels, J. Neurosci. 19 (1999) 1895 – 1911. [8] B.P. Delisle, J. Satin, pH modification of human T-type calcium channel gating, Biophys. J. 78 (2000) 1895 – 1905. [9] K. Talavera, A. Janssens, N. Klugbauer, G. Droogmans, B. Nilius, Extracellular Ca2+ Modulates the effects of protons on gating and conduction properties of the T-type Ca2+ channel a1G (Cav3.1), J. Gen. Physiol. 121 (2003) 511 – 528. [10] J.C. Gomora, J. Murbartia´n, J.M. Arias, J.H. Lee, E. Perez-Reyez, Cloning and expression of the human T-type channel Cav3.3: insights into prepulse facilitation, Biophys. J. 83 (2002) 229 – 241. [11] J.Y. Park, S.W. Jeong, E. Perez-Reyes, J.H. Lee, Modulation of Cav3.2 T-type Ca2+ channels by protein kinase C, FEBS Lett. 547 (2003) 37 – 42. [12] R. Warre, A. Randall, Modulation of the deactivation kinetics of recombinant rat T-type Ca2+ channel by prior inactivation, Neurosci. Lett. 293 (2000) 216 – 220. [13] C.J. Frazier, J.R. Serrano, E.G. George, X. Yu, A. Viswanathan, E. Perez-Reyes, S.W. Jones, Gating kinetics of the a1I T-type calcium channel, J. Gen. Physiol. 118 (2001) 457 – 470.
Influence of external pH on two types of low-voltage-activated calcium currents in primary sensory neurons of rats V.O. Pinchenko*, P.G. Kostyuk, E.P. Kostyuk Department of General Physiology of Nervous System, Bogomoletz Institute of Physiology, The National Academy of Science of Ukraine, Bogomoletz Street 4, Kyiv 01024, Ukraine Received 3 March 2004; received in revised form 29 March 2005; accepted 5 April 2005 Available online 25 April 2005
Abstract The influence of extracellular pH (pHo) on low-voltage-activated calcium channels of acutely isolated DRG neurons of rats was examined using the whole cell patch-clamp technique. It has been found that in the neurons of middle size with capacitance C = 60 T 4.8 pF (mean T S.E., n = 8) extracellular acidification from pHo 7.35 to pHo 6.0 significantly and reversibly decreased LVA calcium current densities by 75 T 3.7%, shifted potential for half-maximal activation to more positive voltages by 18.7 T 0.6 mV with significant reduction of its voltage dependence. The half-maximal potential of steady-state inactivation shifted to more positive voltages by 12.1 T1.7 mV (n = 8) and also became less voltage dependent. Dose – response curves for the dependence of maximum values of LVA currents on external pH in neurons of middle size have midpoint pK a = 6.6 T 0.02 and hill coefficient h = 0.94 T 0.04 (n = 5). In small cells with capacitance C = 26 T 3.6 pF (n = 5), acidosis decreased LVA calcium current densities only by 15.3 T 1.3% and shifted potential for half-maximal activation by 5.5 T 1.0 mV with reduction of its voltage dependence. Half-maximal potential of steady-state inactivation shifted to more positive voltages by 10 T 1.6 mV (n = 4) and also became less voltage dependent. Dose – response curves for the dependence of maximum values of LVA currents on external pH in neurons of small size have midpoint pK a = 7.9 T 0.04 and hill coefficient h = 0.25 T 0.1 (n = 4). These two identified types of LVA currents besides different pH sensitivity demonstrated different kinetic properties. The deactivation of LVA currents with weak pH sensitivity after switching off depolarization to 30 mV had substantially longer decay time than do currents with strong pH sensitivity (s d ¨5 ms vs. 2 ms respectively). It was found that the prolongation of depolarization steps slows the subsequent deactivation of T-type currents in small DRG neurons. Deactivation traces in these neurons were better described by the sum of two exponentials. Thus, we suppose that T-type channels in small DRG neurons are presented mostly by a1I subunit. We suggest that these two types of LVA calcium channels with different sensitivity to external pH can be differently involved in the origin of neuropathic changes. D 2005 Elsevier B.V. All rights reserved. Keywords: Dorsal root ganglion; T-type Ca2+ channels; Acidosis; pH sensitivity; Neuropathy
1. Introduction Diabetic neuropathy is a condition in which hyperglycemia is combined with hypoxia. Both of these factors can induce intracellular and/or extracellular acidification [1]. These pathophysiological changes may contribute to chronic neuropathic pain originating in nociceptors that are subpopulations of the primary sensory neurons (DRG neurons). * Corresponding author. Tel.: +7 38044 2562086; fax: +7 38044 2562093. E-mail address: [email protected] (V.O. Pinchenko). 0304-4165/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.bbagen.2005.04.008
It is well known that the DRG neurons can be divided into three groups according to their somatic sizes [2]. Nociceptors are commonly found among neurons of small and middle sizes and these neurons express both low-voltage-activated (LVA) and high-voltage-activated (HVA) calcium currents [2]. It has been shown [3] that extracellular acidification substantially attenuates inward calcium currents. In contrast to HVA calcium channel studies, only scant information exists about the regulation of LVA calcium channels. The pH sensitivity of native T-type currents has been previously studied only in few cases: in ventrobasal thalamic neurons [3], CA1 hippocampal neurons [4] and cardiac myocytes [5].
2
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
LVA Ca2+ channels start to open when cells are depolarized near 60 mV. They open and inactivate rapidly but their most remarkable kinetic property is that they close (deactivate) much slower than do HVA Ca2+ channels. Earlier, it has been shown [6] that in thalamic neurons, at least two types of LVA calcium channels with different kinetic and pharmacological profiles are expressed. Later, they were classified in 3 subtypes—Cav3.1, 3.2, and 3.3 [7]. Previously, a detailed analysis of the effect of pH changes was made on recombinant a1H [8] and a1G [9] T-type channels. Similar information on pH modulation of a1I subunit is still lacking. Such analysis on native T-type channels might be quite important because of the different location and functional significance of these subtypes. Therefore, we compared such effect on kinetically different T-types channels—with slow and fast inactivation and deactivation.
2. Methods 2.1. Cell preparation In our experiments, we used male 3-month-old Wistar rats. Neurons were freshly isolated from the lumbar dorsal root ganglia (DRG). Rats were decapitated after ether anesthesia, and the ganglia were removed and placed in Tyrode solution. After cleaning, they were incubated for 25 –30 min at 35- in the same solution supplemented with 1 mg/ml protease (Sigma) and 0.5 mg/ml collagenase Type 1 (Worthington). After, the enzymatic treatment ganglia were washed in the enzyme-free solution and dissociated by gentle pipetting. The cell suspension was plated on sterile glass coverslips and incubated in Tyrode solution for 60 min. It was enough to adhere a sufficient number of single cells to the coverslips to perform the recordings. The cells in the recording chamber were placed on the table of Zeiss inverted microscope with phase contrast objective and CCD camera connected to a TV monitor. The cell diameter was taken as a mean of longest and shortest diameters of cell body measured with a ruler on the monitor image. 2.2. Chamber and solutions The experimental chamber of 2 ml volume was continuously perfused (2 ml/min) with external bath solution composed of (mM): 2 CaCl2, 67 choline chloride, 100 tetraethylammonium chloride, 5.6 glucose, 5.3 KCl, 10 Hepes and 3.8 MgCl2. pH was adjusted to the desired value with KOH or HCl. The patch pipettes were filled with an electrode solution composed of (mM): 140 CsCl, 10 Hepes, 10 EGTA, 4 MgATP and 0.1 NaGTP (pH 7.2– 7.3 with CsOH).
Seal resistances >1 GV were achieved using patch pipettes of 1 –2 MV resistances when they were filled with an electrode solution. In all experiments, compensation of series resistance was performed using the standard procedure of the 3900A amplifier (Dagan Corp., Minneapolis, MN, USA). Data were acquired with DIGIDATA 1200 ADAC and amplified with 3900A amplifier using PCLAMP 6.0.3 (Axon Instruments Inc., USA) software, and final data records were analyzed with CLAMPFIT 8.0 program (Axon Instruments Inc., USA) and Microsoft Excel. Current signals were sampled at 20 kHz and filtered at 5 kHz through the interface connected to an AT-compatible computer system which also served as a stimulus generator. 2.4. Voltage protocols and data analysis Activation gating was measured by holding at 80 mV, then applying step to 90 mV for off-line leakage correction followed by V test from 60 mV to +50 mV. The peak current is plotted as a function of V test. We fit the current voltage curves with the Boltzmann equation modified with Golgman– Hodgkin – Katz equation: 0 1 B I Ca ðV Þ ¼ V B @
C G G 1 þ 2 C V1 V V2 V A 1 þ exp 1 þ exp k1 k2 2 3 zF ðV VCa Þ exp 1 6 7 RT 7 6 4 5 zFV 1 exp RT 2 3 zF ðV VL Þ exp 1 6 7 RT 7 ð1Þ þ V G3 6 4 5 zFV 1 exp RT
where G 1, G 2 and G 3 are maximal conductance for LVA, HVA and leakage currents, V 1, V 2, k 1, and k 2 are the midpoints and slope factors for LVA and HVA current activations, V Ca and V L are reversal potentials for calcium and leakage current components, T is absolute temperature, R is universal gas constant, F is the Faraday constant and z is the valence of a permeable ion (equal to 2 for Ca2+ and 1 for leakage currents). Steady-state inactivation is measured by holding at 80 mV, then pre-pulsing from 120 to 10 mV for 2.5 s, followed by a V test, to 30 mV. The peak current of V test is plotted as a function of pre-pulse potential. The data were fit with the Boltzmann equation:
2.3. Electrical recording
I 1 þC ¼ V V1=2 Imax 1 þ exp k
Voltage-clamp recordings using the whole-cell variant of the patch-clamp technique were made at room temperature.
where I/I max is relative current, V is pre-pulse potential, V 1/2 is midpoint potential for complete inactivation, k is the slope
ð2Þ
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
3
of the voltage dependence for inactivation, and C is a constant. The time constants of activation (s m) and inactivation (s h) of the currents were determined from a fit of current traces with the equation: I ðt Þ ¼ I o I½1 expð t=sm Þ2 I expð t=s h Þ þ I c
ð3Þ
where I o is a normalization factor and I c is a constant component. Deactivation time constants (s d) were determined from single-exponential fit of tail currents. Dose – response curves for the pH dependence of LVA currents were built using the following equation: I¼
Imax 1 þ 10ð pK a pH o Þh
ð4Þ
where I max is the maximal asymptotic value of T-current, pK a is pH value at half-maximal current and h is the Hill coefficient. Data are reported as mean T S.E. Student’s t tests on independent groups were used to evaluate P values.
3. Results DRG neurons of small and middle size possess both LVA and HVA calcium channels, with preferential expression of LVA channels in neurons of middle size. We did not use pharmacological tools to separate calcium currents because of the possible pH-dependent effects of the drugs. Thus, we made mathematical separation of LVA and HVA currents using Eq. (1) (see Methods). The example of such mathematical approach is shown in Fig. 1, where the I –V curve for total calcium current in representative cell of middle size was separated into LVA and HVA calcium current components together with leakage current for monovalent cations. Fig. 2 shows the mean I – V curves recorded in cells of middle (A) and small (B) size in control solution with pH 7.35 (filled squares) and under acidosis pH 6.0 (open circles). A striking difference in pH dependence of LVA components for these two types of cells can be clearly seen. The peak of the LVA component in middle size cells (A) was shifted by ¨20 mV in depolarized direction and decreased 5fold under acidosis, while in small cells (B), it decreased only to ¨50% of the initial value and shifted by ¨5 mV rightwards. To estimate the activation parameter changes under acidosis, we recorded I – V curves using 50 ms depolarizing steps with 5 mV increment from V hold = 80 mV. After the mathematical separation of LVA current from HVA current (Eq. (1)), we grouped the cells by the degree of acidosis influence on LVA component and named them as ‘‘strong’’ for those cells in which activation parameters changed substantially comparing with ‘‘weak’’ cells. It turned out that ‘‘strong’’ cells have larger somatic size with mean diameter and membrane capacitance 33 T 1 Am and 60 T 4.8 pF (n = 8), respectively, compared to
Fig. 1. The mathematical separation of experimental I – V curve (open circles) for total calcium current in a representative cell of middle size into LVA current (thin line), HVA current (dotted line) and leakage current (dashed line) using Eq. (1) (see Methods). The sum of these components (bold line) gives the best fit of the experimental I – V curve.
28 T 2 Am and 26 T 3.6 pF (n = 5) for ‘‘weak’’ cells. The influence of acidosis on activation parameters in these two groups of cells and the results of statistical analysis are presented on three panels of Fig. 3. As it can be seen in panel A, the activation midpoint potential was shifted under acidosis by ¨20 mV in the depolarized direction in ‘‘strong’’ cells and only by ¨5 mV in ‘‘weak’’ cells; on panel B, the activation slope factor was also changed to a larger extent in ‘‘strong’’ cells under acidosis (from 3.8 mV/e-fold to 5.2 mV/e-fold) and only from 5.8 mV/efold to 6.2 mV/e-fold in ‘‘weak’’ cells. The most striking difference in pH dependence of these two types of LVA currents on external pH is depicted on Fig. 4. To evaluate the pH dependence of T-currents in cells of middle and small size diameters, we have built the current –voltage curves for calcium currents in following external pHo values: 6.0, 6.5, 7.35, 8.0 and 8.5. Then, the maximum of T-currents at given pH was normalized to that at pH 7.35. Hill curves were plotted through the mean values obtained for several cells grouped by cell size (membrane capacitance). The pK a value and the Hill coefficient h were equal to 6.6 T 0.02 and 0.94 T 0.04 (n = 5) for middle size cells and to 7.9 T 0.04 and 0.25 T 0.1 (n = 4) for small size cells respectively. Thus, T-channels of the middle size cells are strongly influenced by external pH near normal pH 7.35. As the Hill coefficient is close to one and the value of pK a is shifted to the acidic pH value, the strong inhibition of T-currents took place under acidosis. In contrast, T-channels of small size cells exhibit low values of the Hill coefficient and the pK a value that is shifted to alkaline pH, which lead to relatively weak inhibition of the current through these channels under acidosis. Fig. 5 demonstrates the influence of external acidification on steady-state inactivation curves recorded in cells with strong pH sensitivity of LVA currents. External acidification to pHo 6.0 decreases the peak of LVA current, shifts the
4
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
Fig. 2. Mean I – V curves for calcium currents recorded in cells of middle size (n = 5; A) and small size (n = 4; B) in control solution with pH 7.35 (filled squares) and under acidosis pH 6.0 (open circles).
voltage dependence of steady-state inactivation and decreases its slope factor. Panels A and B show the family of currents in representative cell of middle size for pHo 7.35 and 6.0 elicited by a V test of 30 mV, following pre-pulses from 110 to 40 mV for 2.5 s. In panel C, peak currents from 8 middle-sized neurons are normalized to the peak of the maximally available current and relative current is plotted as a function of pre-pulse potential and is fitted with the Boltzmann distribution (solid line). The inactivation midpoint potential V mid is shifted from 69 mV at pH0 7.35 to 57 mV at pHo 6.0, with significant effect on the slope (5.4 mV/e-fold at pHo 7.35 and 7.6 mV/e-fold at pHo 6.0) (n = 8; P < 0.01). Fig. 6 shows the influence of acidification on steady-state inactivation curves recorded in cells with weak pH sensitivity of LVA currents. External acidification to pHo 6.0 decreases the peak of LVA current, shifts the voltage dependence of steady-state inactivation and slightly decreases its slope factor. Panels A and B represents a family of currents in typical cell of small size for pHo 7.35 and 6.0 elicited by the same protocol as in Fig. 5. Panel C
Fig. 3. Bar histograms summarizing acidosis influence on activation parameters in ‘‘strong’’ and ‘‘weak’’ pH-sensitive cells: the activation midpoint potential V mid in ‘‘strong’’ cells is equal to 50 T 1.0 mV in control and 31 T1.2 mV under acidosis; in ‘‘weak’’ cells, this value is equal to 48 T 0.8 mV in control and 43 T 0.7 mV under acidosis (A). The activation slope factor k in ‘‘strong’’ cells is equal to 3.8 T 0.3 mV/e-fold in control and 5.2 T 0.4 mV/e-fold under acidosis; in ‘‘weak’’ cells, this value is equal to 5.8 T 0.1 mV/e-fold in control and 6.2 T 0.1 mV/e-fold under acidosis (B).
shows the mean steady-state inactivation curves for 4 small neurons fitted with a Boltzmann distribution. The inactivation midpoint potential V mid is shifted from 80.5 mV at pHo 7.35 to 70.5 mV at pHo 6.0; also acidosis affects the slope factor, but to less extent than in cells with strong pH
Fig. 4. Dose – response of LVA currents on external pH in middle size cells (open circles) and in small size cells (filled squares). The best fit values of pK a and the Hill coefficient values according to Eq. (4) are shown on a legend.
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
5
Fig. 7. Kinetic properties of two representative traces of strong (bold line) and weak (dashed line) pH-sensitive LVA currents. Fig. 5. The influence of acidification on steady-state inactivation curves recorded in cells with strong pH sensitivity of LVA currents.
sensitivity (4.6 mV/e-fold at pHo 7.35 and 5.0 mV/e-fold at pHo 6.0). The kinetic properties of two representative traces of LVA currents recorded from strong and weak pHsensitive cells are shown in Fig. 7. Holding and test potentials were 80 mV and 30 mV, respectively. A dramatic difference in kinetic of inactivation and deactivation is obvious in these two cells, while the activation times remained almost the same. The voltage and pH dependence of activation and inactivation kinetic parameters for cells with strong and weak pH-sensitive LVA currents are shown in Fig. 8. The time constant for activation s m (panel A) in strong pH-sensitive cells was shifted by ¨20 mV in the direction of depolarization under acidosis, became less voltage dependent and increased approximately 3-fold near
the peak of the I – V curve (at 30 mV). In contrast to this observation, in cells with weak pH sensitivity (panel B), this shift reached only about ¨7 mV under acidosis, which is consistent with charge screening hypothesis. As for inactivation time constants (s h), similar conclusions could be done from the analysis of their pH and voltage dependence in cells with strong and weak pH sensitivity (panels C and D, respectively). In the paper [12], authors investigated a1I subunit expressed in HEK293 cells. They observed that the rate of deactivation at 80 mV was clearly dependent on the level of inactivation induced immediately prior to repolarization, with greater inactivation producing significantly slower deactivation. In our work, we used a similar protocol to investigate the possible influence of the extent of inactivation on the rate of deactivation of T-type currents in small DRG cells. An example of such protocol with current traces is shown on Fig. 9A. Test pulses to 45 mV of 18, 36, 70, 140 and 280 ms duration were applied from holding potential of 80 mV. The duration of repolarization to 80 mV was equal to 20 ms after every test pulse. We found that tail currents in small cells cannot be fitted well with single exponential, thus, we use for this purpose two exponentials. Fig. 9B shows the relationship of weighted average for fast and slow components of deactivation current on the duration of the prior test pulse to 45 mV. Small cells have T-type channels with properties similar to the above-mentioned cloned a1I subunits (strong correlation of deactivation s with test pulse duration).
4. Discussion
Fig. 6. The influence of acidification on steady-state inactivation curves recorded in cells with weak pH sensitivity of LVA currents.
In the present work, we have shown that primary sensory neurons possess at least two distinct types of LVA Ca2+ channels. These channels differ in pH sensitivity and
6
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
Fig. 8. Voltage and pH dependence of activation kinetic parameter (s m) for cells with strong (n = 5; A) and weak (n = 4; B) pH-sensitive LVA currents and inactivation kinetic parameter (s h) for cells with strong (C) and weak (D) pH-sensitive LVA currents at pH 7.35 (filled squares) and pH 6.0 (open circles).
Fig. 9. (A) The example of protocol and current traces used to find the relationship (B) between test-pulse duration and tail current kinetics.
kinetic properties. We have found that small DRG neurons express preferentially LVA Ca2+ channels with weak pH sensitivity while neurons of medium size express LVA Ca2+ channels with strong pH sensitivity. Also, weak pHsensitive channels have slower inactivation and deactivation kinetics than do strong pH-sensitive ones. In order to elucidate the possible a1 subunit composition of these channels, we compared our data with those obtained by other groups on recombinant Ca2+ channels (Cav3.2 [8] and Cav3.1 [9]). It was found that native strong pHsensitive LVA channels are similar to recombinant ones in terms of kinetic properties and influence of external acidification on midpoint potential of activation, but in case of the native channels, a more profound effect on the latter parameter can be seen. To our knowledge, there are no publications concerning the influence of external pH on the Cav3.3 subunit of LVA channels. The inactivation kinetics of weak pH-sensitive LVA Ca2+ channels in our study is similar to that of recombinant Cav3.3 channels [10], but the deactivation kinetics is much slower. Our experimental data indicate that the prolongation of depolarization steps slows subsequent deactivation. This is in good agreement with results obtained by Warre and Randall [12] on a1I subunit. Also, we have found that deactivation traces were better described by the sum of two exponentials. Earlier, Frazier et al. [13] concluded that a single exponential provides an adequate description of tail currents for a1G but not a1I subunits. Thus, we can conclude that the a1I subunit is mostly expressed in the case of small DRG neurons. Taking into account that DRG neurons express mostly Cav3.2 and Cav3.3 mRNA [7], we can suppose that the most probable
V.O. Pinchenko et al. / Biochimica et Biophysica Acta 1724 (2005) 1 – 7
a1 subunit candidates for strong pH-sensitive LVA Ca2+ channels are the Cav3.2, while for weak pH-sensitive LVA Ca2+ channels, the Cav3.3. It has been recently shown that Cav3.2 channels can be also distinguished from other T-type channel subtypes by specific intracellular modulation by protein-kinase C [11], making them very sensitive to intracellular metabolic changes. The possible physiological significance of our findings can be considered in the terms of nociception. The DRG neurons of different size are known to possess different perception modality. The small neurons receive and transmit preferentially nociceptive signals, while neurons of middle and large sizes deal with proprioceptive sensations. Under pathological conditions of acidosis, the different pH sensitivity of LVA channels expressed in these neurons could lead to the shift of total signal modality to the nociceptive side, giving rise to different types of neuropathic changes. Acknowledgement This work was supported by INTAS grant No. 99-01915. References [1] K. Kaila, B.R. Ransom, pH and Brain Function, Wiley-Liss, Inc., 1998. [2] R.S. Scroggs, A.P. Fox, Calcium current variation between acutely isolated adult rat dorsal root ganglion neurons of different size, J. Physiol. (Lond.) 445 (1992) 639 – 658.
7
[3] M.J. Shah, S. Meis, T. Munsch, H.C. Pape, Modulation by extracellular pH of low- and high-voltage-activated calcium currents of rat thalamic relay neurons, J. Neurophysiol. 85 (2001) 1051 – 1058. [4] G.C. Tombaugh, G.G. Somjen, Differential sensitivity to intracellular pH among high- and low-threshold Ca2+ currents in isolated rat CA1 neurons, J. Neurophysiol. 77 (1997) 639 – 653. [5] J. Tytgat, B. Nilius, E. Carmeliet, Modulation of the T-type cardiac Ca2+ channel by changes in proton concentration, J. Gen. Physiol. 96 (1990) 973 – 990. [6] A.N. Tarasenko, P.G. Kostyuk, A.V. Eremin, D.S. Isaev, Two types of low-voltage-activated Ca2+ channels in neurons of rat laterodorsal thalamic nucleus, J. Physiol. (Lond.) 499 (1997) 77 – 86. [7] E.M. Talley, L.L. Cribbs, J.H. Lee, A. Daud, E. Perez-Reyez, D.A. Bayliss, Differential distribution of three members of gene family encoding low voltage-activated (T-type) calcium channels, J. Neurosci. 19 (1999) 1895 – 1911. [8] B.P. Delisle, J. Satin, pH modification of human T-type calcium channel gating, Biophys. J. 78 (2000) 1895 – 1905. [9] K. Talavera, A. Janssens, N. Klugbauer, G. Droogmans, B. Nilius, Extracellular Ca2+ Modulates the effects of protons on gating and conduction properties of the T-type Ca2+ channel a1G (Cav3.1), J. Gen. Physiol. 121 (2003) 511 – 528. [10] J.C. Gomora, J. Murbartia´n, J.M. Arias, J.H. Lee, E. Perez-Reyez, Cloning and expression of the human T-type channel Cav3.3: insights into prepulse facilitation, Biophys. J. 83 (2002) 229 – 241. [11] J.Y. Park, S.W. Jeong, E. Perez-Reyes, J.H. Lee, Modulation of Cav3.2 T-type Ca2+ channels by protein kinase C, FEBS Lett. 547 (2003) 37 – 42. [12] R. Warre, A. Randall, Modulation of the deactivation kinetics of recombinant rat T-type Ca2+ channel by prior inactivation, Neurosci. Lett. 293 (2000) 216 – 220. [13] C.J. Frazier, J.R. Serrano, E.G. George, X. Yu, A. Viswanathan, E. Perez-Reyes, S.W. Jones, Gating kinetics of the a1I T-type calcium channel, J. Gen. Physiol. 118 (2001) 457 – 470.