This study characterized functional ion channels in cultured undifferentiated human mesenchymal stem cells (hMSCs) from bone marrow with whole-cell patch clamp and reverse transcription polymerase chain reaction (RT-PCR) techniques. Three types of outward currents were found in hMSCs, including a noise-like rapidly activating outward current inhibited by the large conductance Ca2+-activated K+ channel (IKCa) blocker iberiotoxin, a transient outward K+ current (Ito) suppressed by 4-aminopyridine (4-AP), and a delayed rectifier K+ current (IKDR)-like ether-à-go-go (eag) K+ channel. In addition, tetrodotoxin-sensitive sodium current (INa.TTX) and nifedipine-sensitive L-type Ca2+ current (ICa.L) were also detected in 29% and 15% hMSCs, respectively. Moreover, RT-PCR revealed the molecular evidence of high levels of mRNA for the functional ionic currents, including human MaxiK for IKCa, Kv4.2 and Kv1.4 for Ito, heag1 for IKDR, hNE-Na for INa.TTX, and CACNAIC for ICa.L. These results demonstrate that multiple functional ion channel currents—that is, IKCa, Ito, heag1, INa.TTX, and ICa.L—are expressed in hMSCs from bone marrow.
Mesenchymal (stromal) stem cells (MSCs) from bone marrow have been recently isolated and expanded in vitro with bone marrow of different species (e.g., mice, rats, and humans); they showed multilineage potential [1–5] to incorporate into a variety of tissues, including bone, cartilage, muscle, lung, and spleen after systemic injection [2, 3, 6] and also to form other kinds of tissue or cells in vitro, such as hepatocytes, cardiomyocytes, and neuronal cells [1, 2, 4]. Animal studies demonstrated that transplantation of MSCs to the infarcted myocardium significantly improved heart function [7–9].
Human MSCs (hMSCs) from bone marrow have shown the potential to differentiate into several types of cells . They were used experimentally in cell therapy for ischemic brain of rat , ischemic myocardium of swine [11, 12], and cardiomyopathy of mouse . It was found that hMSCs appeared to differentiate into cells with a neuron-like phenotype in brain and improve functional performance of the apoplectic animal  or into cardiomyocytes in myocardium and improve heart contractile function [11–13]. The hMSCs are characterized with high expansion potential, genetic stability, reproducible characteristics in widely dispersed laboratories, compatibility with tissue engineering, and potential to enhance repair in many vital tissues [3, 14–18]. In addition, hMSCs were used as a gene delivery system to deliver therapeutic genes ; for example, the cardiac pacemaker gene mHCN2 was transfected into hMSCs to create cardiac pacemakers .
Ion channels are extensively expressed in different types of cells, and they have important roles in maintaining physiological homeostasis. However, expression of ion channels is not well documented in hMSCs. A recent report described that large-conductance Ca2+-activated K+ current (IKCa), L-type Ca2+ current, and slow K+ current (Is) were present in hMSCs . The present study demonstrated that, in addition to the ionic currents reported previously, three more ionic currents were coexpressed in undifferentiated hMSCs. Properties and molecular biological basis of these ion channels were characterized with whole-cell patch and reverse transcription polymerase chain reaction (RT-PCR) techniques.
Materials and Methods
Normal hMSCs derived from bone marrow were purchased (passage 3) from Cambrex Bio Science (Baltimore, MD; http://www.cambrex.com), which are positive for CD105, CD166, CD29, and CD44 and negative for CD14, CD34, and CD45. The cells were cultured as monolayers in mesenchymal stem cell growth medium (MSCGM; Cambrex Bio Science) containing 10% fetal bovine serum and antibiotics (50 U/ml penicillin and 50 μg/ml streptomycin; Invitrogene, Carlsbad, CA; http://www.invitrogen.com) at 37°C in a humidified atmosphere of 95% air and 5% CO2. When the cells had grown in culture flasks to 70%–80% confluence, they were detached with trypsin/EDTA. After centrifugation at 120 g for 5 minutes, cells were either used for the extraction of RNA or suspended in medium for continuous culture and ionic current recording. For ionic current study, cells were transferred to a cell chamber mounted on the stage of the inverted microscope (DM IL; Leica; http://www.leica.com ) for 15–20 minutes and allowed to attach to the bottom of the cell chamber. Subsequently, the cells were superfused with normal Tyrode solution (1.5 ml/ min). The cells we used for RNA extraction and electro-physiological study were from passages 4 to 8.
The Tyrode solution contained 136 mM NaCl, 5.4 mM KCl, 1.0 mM MgCl2, 1.8 mM CaCl2, 0.33 mM NaH2PO4, 10 mM glucose, and 10 mM HEPES; the pH was adjusted to 7.4 with NaOH. The pipette solution contained 20 mM KCl, 110 mM K-aspartate, 1.0 mM MgCl2, 10 mM HEPES, 0.05 mM EGTA (or 5wherespecified), 0.1mMGTP, 5.0mM Na2-phosphocreatine, and 5.0 mM Mg2-ATP; the pH was adjusted to 7.2 with KOH. K+ in superfusion and pipette solutions were replaced by equimolar Cs+ when K+-free conditions were applied for recording sodium current (INa) or L-type Ca2+ current (ICa.L). The experiments were conducted at room temperature (21°C–22°C).
Data Acquisition and Analysis
The whole-cell patch-clamp technique was used. Borosilicate glass electrodes (1.2 mm outside diameter [OD]) were pulled with a Brown-Flaming puller (model P-97; Sutter Instrument Co., Novato, CA; http://www.sutter.com), and had tip resistances of 2 to 3 MΩ when filled with pipette solution. The tip potentials were compensated before the pipette touched the cell. After a giga-seal was obtained by negative suction, the cell membrane was ruptured by gentle suction to establish the whole-cell configuration. Data were acquired with an EPC9 amplifier (HEKA Elektronik Lambrecht/Pfalz, Germany; http://www.heka.com). Membrane currents were low-pass filtered at 2 kHz, then stored on the hard disk of an IBM-compatible computer.
Reverse Transcription Polymerase Chain Reaction
Total RNA of hMSCs (from passages 4–8) was isolated using the TRIzol method and further treated with DNase I (both from Invitrogen). Reverse transcription (RT) was performed with the RT system (Promega Corp., Madison, WI; http://promega.com) protocol in a 20-μ1 reaction mixture. RNA (1μg) was used in the reaction, and a combination of oligo(dT) and random hexamer primers was used for the initiation of cDNA synthesis. After this RT procedure, the reaction mixture (cDNA) was used for polymerase chain reaction (PCR). The cDNA was replaced by sterile nuclease-free water for negative control in each pair of primers, and we did not find any significant band in the negative control.
The forward and reverse PCR oligonucleotide primers chosen to amplify the cDNA are listed in Table 1. PCR was performed by a Promega PCR system with Taq polymerase and accompanying buffers. The cDNA at 2-μ1 aliquots was amplified by a DNA thermal cycler (Mycycler; Bio-Rad Laboratories, Hercules, CA; http://www.bio-rad.com) in a 25-μ1 reaction mixture containing 1.0× thermophilic DNA polymerase reaction buffer, 1.25 mM MgCl2, 0.2 mM of each deoxynucleotide triphosphate (dNTP), 0.6 μM of each forward and reverse primer, and 1.0 U of Taq polymerase under the following conditions: the mixture was annealed at 50°C–61°C (1 minute), extended at 72°C (2 minutes), and denatured at 95°C (45 seconds) for 30–35 cycles. This was followed by a final extension at 72°C (10 minutes) to ensure complete product extension. The PCR products were electrophoresed through a 1% agarose gel, and amplified cDNA bands were visualized by ethidium bromide staining. The bands imaged by Chemi-Genius Bio Imaging System were analyzed via GeneTools software (both Syngene, Cambridge, UK; http://www.syngene.com).
Table Table 1.. Oligonucleotide sequences of primers used for RT-PCR
MaxiK, human large-conductance, voltage- and calcium-activated K+ channel; hKv, human voltage-gated K+ channel; heag, human either-à-go-go K+ channel; hNE-Na, tetrodotoxin-sensitive voltage-activated Na+ channel from human neuroendocrine; SCN5A, human cardiac tetrodotoxin-insensitive voltage-dependent Na+ channel, a-subunit; CACNA1C, human voltage-dependent L-type Ca2+ channel, alpha 1C subunit; CACNA1G, human voltage-dependent T-type Ca2+ channel, alpha 1G subunit; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.
Forward primer (5′-3′)
Reverse primer (5′-3′)
Nonlinear curve-fitting programs (Pulsfit or Sigmaplot; SPSS, Chicago, IL; http://www.spss.com) were used. Results are presented as means ± SEM. Paired and unpaired Student's t-tests were used as appropriate to evaluate the statistical significance of differences between two group means, and analysis of variance (ANOVA) was used for multiple groups. Values of p < .05 were considered to indicate statistical significance.
Families of Ion Channel Currents
Figure 1 illustrates families of membrane currents elicited by 300-ms voltage steps to between −60 and +60 mV from a holding potential of −80 mV, as shown in the inset of Figure 1B. Figure 1A displays two components of ionic currents activated by the depolarizing voltage steps in a representative hMSC. One component showed a gradual activating current at potentials between −20 and +30 mV—that is, a delayed rectifier K+ current (IKDR)—and another was a rapidly activating current with noisy oscillation at +40 to +60 mV, similar to voltage-activated and Ca2+-activated K+ current (IKCa) reported recently by Heubach and colleagues . Figure 1B shows a transient outward current, similar to Ca2+-resistent transient outward K+ current (Ito) in cardiac and neuronal cells [21–23], coexisting with the noise-like IKCa in another hMSC. Figure 1C displays current traces recorded in another typical experiment, showing three types of currents: an inward current, followed by IKDR and noise-like IKCa (at positive potentials of +50 and +60 mV). Almost all of the hMSCs investigated (149 of 154 cells, from different dishes of passages 4 to 8) demonstrated outward currents (mostly IKCa at more positive potentials) activated by the voltage protocols, while Ito was found in 8% (12 of 149 cells) of the hMSCs. The inward current was coexistent with outward currents (i.e., IKCa, IKDR, or Ito) in 29% of the hMSCs (43 out of 149 cells). The hMSCs studied had resting membrane potentials between −12 and −42 mV. The mean value of the membrane capacitance was 59.7 ± 12.1 pF. Based on the calculation of membrane capacitive charge (1–1.3 pF/μm2) , the average surface area of hMSCs would be 59.7–77.6 μm2. No differences in channel type expression or ion current density were observed in the cells from different passages (4–8).
It is well known that the inward current elicited by depolarizarion voltage steps is carried by Na+ or Ca2+. To study the nature of the inward current, the sodium channel blocker tetrodotoxin (TTX) was employed in seven hMSCs with inward current. Figure 2A illustrates current traces recorded in a typical experiment with K+ pipette solution in normal Tyrode solution. A significant inward current was followed by gradually activating IKDR. TTX at 100 nM abolished the inward current, and the effect recovered after drug washout for 5 minutes, suggesting that the inward current may be TTX-sensitive INa.
Figure 2B displays INa traces recorded under K+-free conditions in a representative hMSC with 30-ms voltage steps (as shown in the inset) in the absence and presence of 50 nM TTX. TTX reversibly suppressed INa. TTX-sensitive INa was found in 6 out of 20 cells, and no inward current was observed in the remaining 14 out of 20 cells. Figure 2C shows the current-voltage (I-V) relationships of INa during control, after the application of 50 nM TTX, and after washout of the drug for 5 minutes in cells with INa. INa peaked at −15 mV with a threshold potential of −40 mV. TTX at 50 nM inhibited INa (measured at −15 mV) to 2.1 ± 0.3 pA/pF from 6.3 ± 0.7 pA/pF of control (p < .01), and recovered to 5.8 ± 0.8 pA/pF after the drug washout for 5 minutes. These results indicate that TTX-sensitive INa (INa.TTX) is present in hMSCs.
It was reported that ICa.L was present in a small population of hMSCs [20, 25]. We determined ICa.L with 200-ms voltage steps to between −40 and +50 mV from a holding potential of −50 mV (to inactivate INa.TTX), because the coexistence of INa.TTX and nifedipine-sensitive ICa.L was observed at a holding potential of −80 mV in a few of the hMSCs (Fig. 3A). At the holding potential of −50 mV, we found that ICa.L (sensitive to inhibition by the ICa.L blocker nifedipine) was present in 4 out of 27 cells. Figure 3B shows ICa.L traces recorded from a representative cell with the protocol shown in the inset during control (left panel) and after the application of 5 μM nifedipine (right panel). The I-V relationship of ICa.L displays the current peaked at 0 mV with density of 0.8 ± 0.3 pA/pF in the control and 0.3 ± 0.2 pA/pF after 5 μM nifedipine (Fig. 3C Figure 3., n = 4, p < .01). The results suggest that dihydropyridine-sensitive ICa.L, as recently reported [20, 25], is present in a small population of hMSCs.
Effect of Iberiotoxin on IKCA
Figure 4 shows the effect of iberiotoxin, a selective blocker of large-conductance IKCa (MaxiK) channels, on IKCa in hMSCs. Iberiotoxin (100 nM; Alomone Labs, Jerusalem, Israel; http://www.alomone.com) substantially inhibited IKCa without affecting inward current and IKDR. Membrane current measured at +60 mV was reduced to 31.5% ± 9.2% of control in a total of five cells. Iberiotoxin-sensitive current showed significant outward rectification (typical of large-conductance IKCa), consistent with the recent report .
Properties of Ito
Ito was detected in a small population (8%) of hMSCs. Figure 5A displays Ito traces recorded in a representative hMSC under control conditions and after the application of 3 mM 4-aminopyridine (4-AP). Ito was substantially inhibited, while noise-like IKCa was slightly suppressed by 4-AP. Ito at +50 mV was inhibited to 0.5 ± 0.2 pA/pF (by 86%) from 3.6 ± 1.1 pA/pF of the control (n = 6). Figure 5B shows voltage-dependent inactivation of Ito assessed by step potential at +50 mV after 1,000-ms variable conditioning potentials (as shown in the inset). The Ito inactivation curve (bottom panel) was obtained by plotting relative availability of Ito as a function of the conditioning potential. The voltage dependence of inactivation (i.e., availability, I/Imax) was fit to the Boltzmann function with half availability (V0.5) of −42.1 ± 2.7 mV and a slope factor of 12.9 ± 1.5 (n = 6).
Figure 5C illustrates time-dependent recovery of Ito from inactivation that was studied with a paired-pulse protocol, as shown in the inset. Ito recovery was complete within 700 ms and well fitted by a monoexponential function (bottom panel) with the time constant (τ) of 175.8 ± 45.7 ms (n = 6). These results indicate that the properties of Ito in hMSCs—that is, 4-AP sensitivity, voltage-dependent inactivation, and time-dependent recovery from inactivation—are similar to those observed in neuronal, smooth muscle, and cardiac cells [21–23, 26].
Properties of IKDR
IKDR was determined in hMSCs under conditions of high concentration of EGTA (5 mM) in pipette solution, along with 200 μM Cd2+ and 100 nM TTX in superfusion solution to inhibit IKCa, ICa.L, and INa. Figure 6 shows voltage-dependent IKDR gradually activated upon 300-ms voltage steps (as shown in the inset), with a significant tail current at −30 mV. IKDR had a linear I-V relationship with threshold potential of −20 mV. The activation variable (g/gmax) was determined from the I-V relationship of IKDR tail current for each cell and fitted to the Boltzmann equation to obtain voltage for half-activation (V0.5) and slope factor (S). Mean V0.5 for activation of IKDR was +8.9 ± 1.1 mV, and S was 14.6 ± 1.4 (n = 11).
Figure 6E displays IKDR traces elicited by long depolarization (5-second) voltage steps to between −60 and +60 from −80 mV in a typical experiment, showing that IKDR does not inactivate after its activation. Similar results were obtained in the other six cells.
Figure 7A shows how the activation kinetics of IKDR changed with alteration of holding potentials. The current reached a steady-state level during 300-ms depolarization with a holding potential of −50 mV, but not with a holding potential of −80 mV, suggesting that the activation process of the current depends on the holding potential. Figure 7B illustrates the IKDR traces and protocol used to evaluate activation time constant. The IKDR activation rate gradually increased as the conditioning potential became more positive. The activation process of IKDR was fit to a monoexponential function. Figure 7C displays mean values of time constants of the current activation at variable conditioning potentials. The time constant became smaller as the conditioning potential was increased to more positive potentials, indicating a faster activation of IKDR at more positive conditioning potentials. These properties are similar to those observed in cloned voltage-gated ether à go-go (eag) K+ channels from rat, mouse, and human [27–30].
The activation of eag K+ channels is dependent not only on conditioning potential but also on extracellular Mg2+ concentration [27, 31, 32]. To study the possible contribution of eag K+ channels to IKDR in hMSCs, we examined whether alteration of extracellular Mg2+ concentration would change activation kinetics of IKDR. Figure 8A displays IKDR activation kinetics regulated by extracellular Mg2+. The activation process of IKDR became slower as the concentration of extracellular Mg2+ was elevated from 0 to 0.2, 0.5, and 5 mM. Mean values of time constant for IKDR activation are illustrated in Figure 8B. The time constant was significantly increased when the extracellular Mg2+ was elevated to 0.2, 1, and 5 from 0 mM at different conditioning voltages (p < .01, n = 7). These features—noninactivation, voltage-dependent activation, and extracellular Mg2+-dependent activation—indicate that IKDR may be contributed by eag K+ channels in hMSCs.
Figure 9 displays the effect of tetraethylammonium (TEA) on the eag K+ channel in hMSCs. TEA at 5 mM reversibly suppressed eag K+ current (Fig. 9A). The I-V relationships of eag K+ current in the absence and presence of 5 and 10 mM TEA are illustrated in Figure 9B (n = 7). TEA significantly inhibited the eag K+ current at test potentials from −10 to +60 mV (p < .05 or p < .01 vs. control). The mean value of the concentration-dependent response of the eag K+ current to TEA is shown in Figure 9C. The concentration giving 50% inhibition (IC50) of eag K+ current by TEA was obtained by fitting the concentration response curve with the Hill equation. IC50 was 2.4 mM, and the Hill coefficient was 0.99.
Message RNA Expression of Functional Ion Channel Currents
To study molecular identity of the functional ionic currents observed, we examined related gene expression in hMSCs with RT-PCR using the specific primers shown in Table 1. Figure 10A displays the mRNA expression for ion channel α-subunits related to functional outward and inward currents. High mRNA levels of MaxiK (responsible for iberiotoxin-sensitive IKCa), hKv1.4 and hKv4.2 (responsible for 4-AP-sensitive Ito), heag1 (responsible for IKDR), hNE-Na (responsible for INa.TTX), and CACNA1C (responsible for ICa.L) were detected in hMSCs. The relative levels of the specific mRNA to the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) are summarized in Figure 10C. These results provide the molecular basis for the functional ionic currents (i.e., IKCa, Ito, heag1, INa.TTX, and ICa.L) observed in hMSCs.
The results from these observations further confirmed that large-conductance IKCa and ICa.L were present in hMSCs. The novel finding obtained from the study reported here was that, in addition to the IKCa and ICa.L reported previously, INa.TTX, Ito, and heag1 K+ current were expressed in undifferentiated hMSCs, and selective RT-PCR screening provided evidence for mRNA species that encode for each current type.
Previous Studies on Ionic Currents in hMSCs
Although hMSCs were used for a number of years in the investigation of cell therapy and differentiation [1, 6, 33–35], information concerning ion channel expression has not been well documented. Kawano and colleagues [25, 36] first studied ionic homeostasis in hMSCs and demonstrated that Ca2+ oscillations regulated by Na+-Ca2+ exchanger and plasma membrane Ca2+ pump might induce fluctuations of membrane current and potentials in hMSCs. In addition, they found that IKCa was present in most of the cells with a conductance of ∼170 pS, and nifedipine-sensitive ICa.L in a small population (15 %) of hMSCs . These two types of currents were also observed by Heubach et al. . Moreover, this group reported a slowly activating K+ current (Is) in hMSCs, which is different from rapidly activating IKCa, and described a high expression level of MaxiK mRNA responsible for IKCa and α1C mRNA of the L-type Ca2+ channel. Significant expression of the ion channel mRNA was also detected in hMSCs, including Kv1.4, Kv4.2, Kv4.3, HCN2, and others, but no functional channel current was recorded in their observation . Our study provides novel information that, in additional to IKCa and ICa.L, three more functional ion channel currents (i.e., Ito, INa.TTX, and heag1) are expressed in undifferentiated hMSCs.
Characteristics of Ion Channel Currents in hMSCs
We first demonstrated here that INa was recorded in about 29% of hMSCs. The current was highly sensitive to blockage by TTX (Fig. 4). Therefore, the current would be TTX-sensitive INa.TTX. The evidence for a high level of mRNA expression of the TTX-sensitive Na+ channel gene hNE-Na (but not SCN5A) further demonstrated that INa.TTX was present in hMSCs (Fig. 10).
We recorded ICa.L in about 15% of hMSCs and found that the current was sensitive to inhibition by nifedipine and that CACNA1C mRNA of the L-type Ca2+ channel (but not the T-type Ca2+ channel CACNA1G) was expressed in hMSCs (Fig. 10). These results further support the observation that ICa.L is present in a small population of hMSCs by other groups [20, 36].
Ito in hMSCs was sensitive to inhibition by 4-AP (Fig. 3A) and showed voltage-dependent inactivation and time-dependent recovery from inactivation (Fig. 3B, 3C). These properties are similar to those of Ito in neuronal, smooth muscle, and cardiac cells [21–23, 26]. It is generally believed that Ito is encoded by Kv1.4, Kv4.2, or Kv4.3. High expression levels of hKv4.2 and hKv1.4 mRNA were found in hMSCs (Fig. 10), suggesting that Ito in hMSCs may be encoded by hKv4.2 and hKv1.4. However, a possible contribution from Kv4.3 could not be excluded, since this isoform was detectable in this study and a high level of expression of Kv4.3 was observed in hMSCs by Heubach et al. .
The study reported here demonstrated that IKCa was coexpressed with other ionic currents in hMSCs and showed characteristics with (1) rapidly activating noisy current traces upon depolarization to more positive potentials, (2) sensitivity to the specific large-conductance Ca2+-activated channel blocker iberiotoxin (Figs. 1 and 2), and (3) a high expression level of MaxiK mRNA (Fig. 10). The observation for the presence of a voltage-activated and Ca2+-activated large-conductance K+ channel in hMSCs is consistent with the report by Heubach et al. .
Significant IKDR was recorded using high EGTA in pipette solution and Cd2+ in superfusion solution to inhibit the activation of IKCa. Resting membrane potential was relatively negative (−25 ∼ −42 mV) in cells with significant IKDR. IKDR was slowly activated with depolarization voltage steps and showed noninactivation (Fig. 6E). More negative conditioning potentials and higher concentration of extracellular Mg2+ slowed the activation of IKDR (Figs. 7 and 8). This property is believed to be the most notable characteristic of eag K+ channels [27, 37], suggesting that IKDR is encoded by the heag gene. It has been demonstrated that two types of heag K+ channels—heag1 and heag2—are identified in human tumor cells [27, 37]. RT-PCR revealed a high level expression of mRNA for the heag1 K+ channel in hMSCs (Fig. 10), indicating that IKDR may be contributed by the heag1 K+ channel.
Heubach et al.  demonstrated that cells with significant Is had relatively negative resting potential and that TEA significantly inhibited the current with IC50 about 2 mM in hMSCs. These properties are similar to those we observed in IKDR. Therefore, the Is is most likely to be IKDR and contributed by the heag1 K+ channel.
Potential Significance and Limitation
It is well known that ion channels have important roles in maintaining physiological homeostasis in different types of cells. Therefore, the multiple expression of ion channels would suggest possible differential roles of these channels in the cellular physiological activity of hMSCs. MaxiK channels (i.e., large-conductance IKCa) are usually believed to be sensors of intracellular Ca2+ and are found to regulate membrane potential in an intracellular Ca2+-dependent manner in hMSCs . Modulation of intracellular Ca2+ depended on several mechanisms, and voltage-operated Ca2+ current did not contribute much [25, 36]. Generally, most of the proliferating cells are believed to lack voltage-gated Ca2+ channels . Our and others' observations showed that hMSCs demonstrated a small dihydropyridine-sensitive ICa.L in a small population of cells [20, 25, 36]. In addition, INa.TTX was also found in about 30% of hMSCs (Fig. 2), but whether both ICa.L and INa.TTX are necessary for the differentiation of hMSCs into excitable cells [10–12] remains to be studied.
Voltage-gated K+ currents are found to modulate the progression through the cell cycle in proliferating cells . Mammalian eag K+ channels have been demonstrated in several species including rat , mouse , bovine , and human . Heag K+ channels were found to hyperpolarize cell membranes and participate in cell proliferation in human breast cancer cells, and inhibition of heag K+ channels induced a depolarization of membrane potential and arrested cells in the early G1 phase . In hMSCs with significant heag1 K+ current (i.e., IKDR), resting membrane potential was relatively negative in our study, which may imply a contribution by the heag1 K+ channel. The literature for the effect of 4-AP–sensitive Ito on cell proliferation is limited; however, the reports from Czarnecki and colleagues [41, 42] demonstrated that Ito might play a part in GH3 cell proliferation processes. Whether and how the heag1 K+ channel and Ito would participate in the regulation of proliferation or differentiation of hMSCs requires additional experimental study in the future.
Differential expression of the inward and outward currents in individual cells suggests that hMSCs may be inhomogeneous. Possible reasons for the heterogeneity may be related to the fact that the cells investigated are not from a homogeneous population of hMSCs because they are cells at different phases of the cell cycle. It was reported that ion channel expression changed with cell cycle progression [40, 43, 44]. In addition, fractions of more or committed progenitor cells would also affect the expression of ion channel patterns [17, 20]. Some differences of electrophysiological properties in hMSCs between this observation and previous reports  are likely related, at least in part, to the reasons described above or to slight differences in cell culture conditions .
In summary, the present study demonstrates for the first time that five distinct ion channel currents are expressed in undifferentiated hMSCs, including three types of outward currents (IKCa, Ito, and IKDR or heag1), and two types of inward currents (INa.TTX and ICa.L), which provides a strong basis for a study of physiological roles of these functional ionic currents in proliferation and differentiation of hMSCs in the future for a further understanding of human biology.
This study was supported by grant no. HKU 7347/03M from the Research Grant Council of Hong Kong. We thank Professor T. M. Wong in the Department of Physiology for his support.