| Background and AimsThe present therapy for bradyarrhythmias consists of the use of electronic pacemakersto sustain the heart rate. However, such devices are not optimal because of the high cost,limited battery life, the lack of a biological response to enable adaptation toneurotransmitter changes following changes in physiological conditions, and undesirablecomplications, such as hemorrhage and bacterial infection. With improved understanding ofthe genetic determinants of ion channel function in pacemaker cells, the attention of anincreasing number of researchers is turning to the field of biological pacing, which couldreplace the traditional electronic pacemaker in the treatment of bradyarrhythmias.The funny current (If) of sino-atrial node cells, which plays a key role in the process ofpacemaker generation and in the modulation of autonomic transmitters for heart rate, isgenerated by hyperpolarization-activated cyclic nucleotide-gated (HCN) channels. Thus far,proof-of-concept data have been generated in recent experiments, in which transplantedmesenchymal stem cells (MSCs) carrying HCN genes generated pacemaker activity inmodels of complete heart block. However, There are some key issues still to be unclear. Onthe one hand, it was shown in the above studies that the beating rate ranged from about40to50bpm, which was significantly lower than that of cardiomyocytes cocultured withMSCs transfected with HCN in vitro. The specific mechanism underlying this phenomenonremains fragmented. On the other hand, MSCs have the potential for multi-directionaldifferentiation, which raised concerns about whether MSCs used only as a gene deliverysystem could acquire the functional attributes of excitable cells via the hostmicroenvironment-mediated induction of differentiation. The above issues is crucial for thetransplanted cells to play a stable and long-lasting pacing function in vivo.Nevertheless, due to the lack of direct experimental access to the implanted cells in situ,current knowledge of the changes in electrophysiological characteristics and the mechanisms underlying the pacemaker activity of engrafted HCN gene-transfected MSCs invivo remains unclear. In previous studies, the functional examination of engrafted MSCsrelied chiefly on immunohistochemical endpoints or data from electrocardiograms (ECG)and optical action potentials(APs) mapping. But the expression of a cardiac myocyte-likephenotype is not equivalent to obtaining cardiac myocyte-like function, and the noninvasiveexamination is subject to their restricted spatial resolution and being difficult to confirm thegeneration of APs.Because ion channels form the basis for engrafted cells to generate pacemaker functionin vivo and because the patch-clamp technique is the most fundamental and effectivemethod for detecting ion channel currents, we performed in situ patch-clamp recording ofengrafted MSCs using vital ventricular slices and investigated the related ionic currents oftheir pacing action, as well as their gap junctional communication with host cardiomyocytes,to provide more insight into electrophysiological properties of engrafted cells and itsinteraction with endogenous myocytes. To the best of our knowledge, this is the first reportdescribing in situ patch-clamp recording on engrafted cells to examine their ioncharacteristics.Method1. Bone marrow specimens were extracted. MSCs were isolated by gradientcentrifugation and its character of adherence to culture plates, and further amplified andpurified in vitro.2. The lentiviral vector pLenti6.3-IRES2-EGFP (which is designated as LV-EGFP) andpLenti6.3-mHCN4-IRES2-EGFP(which is designated as LV-HCN4-EGFP) wereconstructed and produced. Passage-3rMSCs were transfected with LV-HCN4-EGFP orLV–EGFP at multiplicities of infection (MOIs) of10. After being transfected for5~7days,rMSCs were tested for the expression of EGFP and mHCN4protein by usingImmunohistochemical method and patch-clamp technique.3. MSCs were delivered into host heart to build a cell transplantation modle, and aneffective method was established to prepare viable ventricular slices to meet therequirements for patch-clamp detection.4. SD rats were divided into the following two groups: i) experimental group, were transplanted with mHCN4-EGFP-transfected rMSCs, and ii) in vivo control group, weretransplanted with EGFP-transfected rMSCs. The rats were sacrificed at4weeks after cell implantation and the engrafted MSCs were investigated using in situpatch-clamp recording. In addition, as in vitro controls, the recording was also performed inthe following two groups: i) freshly mHCN4-transfected rMSCs before prolonged culture,and ii) after parallel culture for4weeks in vitro. Throughout the report, these four groupsare designated as follows: experimental group, in vivo control group, in vitro(i) and invitro(ii), respectively. The pacemaker current (If) of the cells in each group were detectedusing patch-clamp technique, and the current characteristics of each group were furthercompared and analyzed. In addition, action potential and other current associated with thepacing function, such as sodium current and L-type calcium current, also were detected.5. Gap junctional intercellular communication between engrafted rMSCs and residentcardiomyocytes was determined by the use of Lucifer yellow dye on the basis of the wholecell patch clamp sealing.6. Myocardial tissue adjacent to the sections detected using the patch-clamp techniquewas fixed in4%paraformaldehyde, cryopreserved overnight, and embedded in paraffin.Serial sections (5-μm-thick) were cut from the tissue. The immunofluorescence method wasused for detection of HCN4and EGFP protein expression in the transplanted cells, while theimmunohistochemical method was used for the detection of expression and distribution ofconnexin43(CX43), as well as the presence of IgG antibody and CD3+T cells in thetransplantation area.Results1. In this experiment, purified MSCs express the surface molecules CD29and CD44inthe absence of CD34, CD45, which was assessed by fluorescence-activated cell sorteranalysis, and have the capacity for differentiation to osteoblasts, adipocytes, and myocardialcells in vitro. The above characteristics are in line with the currently accepted standard ofMSCs.2. At a MOI of10, the transfection efficiencies of rMSCs were67.0±6.6%(n=5),while MSCs morphology and activity was unaffected. EGFP-transfected rMSCsdemonstrated EGFP but not HCN4protein expression, whereas mHCN4-EGFP-transfected rMSCs expressed both EGFP and HCN4protein. The results of patch clamp recording showthat Ifwas elicited from mHCN4-EGFP-transfected rMSCs, but not from EGFP-transfectedrMSCs.3. Based on the improved method of preparation of ventricular slices, we obtainedviable ventricular slices with clean surfaces, low fluorescence background and maintainedthe structure integrity of cells. Clusters of transplanted rMSCs distributed within theinterstitial compartment of cardiomyocytes remained round-shaped, and a small number ofthe cells exhibited a short spindle, arranged in parallel with the myocardial cells.Futhermore, membrane currents of myocardial cells and MSCs were successfully recordedby using patch-clamp recording of ventricular slices, which suggested that the viableventricular slices were fully able to meet the requirements for patch-clamp detection.4. Patch-clamp experiments results suggested that the allograftedmHCN4-EGFP-transfected rMSCs of experimental group survived in the host heart for over4weeks, that they expressed Ifwith a similar amplitude but with a more negative activationcompared with parallel mHCN4-transfected rMSCs cultured in vitro(i) and in vitro(ii),while the EGFP-transfected rMSCs of in vivo control group did not express Ifcurrents.(1) There were trends toward larger current densities (at-140mV) inmHCN4-EGFP-transfected rMSCs in situ (-151.7±26.1pA/pF, n=14) compared with invitro(i) cells (-41.6±7.7pA/pF,n=16) and in vitro(ii) cells (-31.7±4.0pA/pF, n=15) andtoward smaller membrane capacities in situ (6.8±1.2pF, n=14) compared with in vitro(i)(25.0±5.6pF,n=16) and in vitro(ii)(32.4±4.8pF, n=15), and for any two of the three sets ofdata, the differences were significant (P<0.05for each). However, there were no significantdifferences in current amplitudes (at-140mV) among the three groups (in situ,-1007.4±132.3pA, n=14; in vitro(i),-1012.5±187.3pA, n=16; in vitro(ii),-1025.5±200.9pA, n=15; P=0.957).(2) All of the mHCN4-EGFP-transfected rMSCs studied in vitro exhibited thresholdvalues between-40and-50mV [in vitro(i), mean value-40.6±4.4mV (n=20), vs. invitro(ii), mean value-41.3±5.2mV (n=22); P=0.181]. In comparison, in20engraftedmHCN4-EGFP-transfected rMSCs studied in situ, the threshold of Ifactivation varied from-40to-90mV (mean value-46.7±5.9mV), and both of the differences reached statisticalsignificance (P<0.05for each) when compared with in vitro(i) and in vitro(ii), respectively. (3) The half-maximal activation voltage for Ifcurrent in situ (-107.8±14.6mV,n=14)was more negative, respectively, compared with that of in vitro(i)(-94.8±7.8mV, n=16;P<0.05) and that of in vitro(ii)(-96.4±6.6mV, n=15; P<0.05). However, the differencebetween the groups of in vitro(i) and in vitro(ii) did not reach significance (P=0.553).(4) Similarly, the slope values of the voltage dependences in situ (-13.7±1.7mV,n=14) was different from that in vitro(i)(-11.5±1.8mV, n=16; P<0.05) and that invitro(ii)(-11.2±2.2mV, n=15; P<0.05), but there was also no significant difference betweenthe two groups in vitro(P=0.717).(5) At-140mV, the time constant for Ifactivation was509.8±66.6ms (n=6) for insitu cells compared with405.6±64.5(n=8) for in vitro(i) cells and427.5±62.2ms (n=7) forin vitro(ii) cells, but the difference among the three groups did not reach statisticalsignificance(P=0.051).(6) The reversal potentials were-32.3±3.3mV (n=8) in situ and-29.5±3.9mV (n=11)in vitro(i) and-30.9±3.0mV (n=10) in vitro(ii), respectively. There was no significantdifference among the three groups (P=0.288).(7) In the presence of3μM isoproterenol, the half-maximum activation voltage inengrafted MSCs was shifted approximately9.7mV (from-106.5±9.5mV to-96.8±8.7mV,n=4; P<0.05) in the positive direction, whereas its slope was not significantly modified(from-12.8±2.1to-12.9±1.8mV).5. Neither sodium currents nor calcium currents were elicited inmHCN4-EGFP-transfected rMSCs in situ (n=10),in vitro(i)(n=12) and in vitro(ii)(n=15).Similarly, an action potential was not evoked in all groups of cells in the current-clampconfiguration. The results were similar in EGFP-transfected rMSCs in situ (n=10). Inaddition, in the fluorescent dye transfer experiments conducted on the basis of the wholecell patch clamp recording, dye did not transfer from those investigated rMSCs (n=6) to theadjacent myocardial cells or other transplanted cells.6. Immunohistochemical results show that only a few rMSCs in vivo expressedconnexin43in a immature manner, which was randomly distributed at the contact interfacesbetween engrafted MSCs and host myocytes in a punctate pattern and without properalignment, significantly different from that of myocardial cells (confined to the intercalateddisc). In addition, there was no binding of rat IgG to the surface of rMSCs and no significant infiltration by CD3+T lymphocytes in the area of cell transplantation.Conclusions1. we established a improved method of preparation of ventricular slices, andsuccessfully performed in situ patch-clamp recording of engrafted MSCs using vitalventricular slices and investigated the related ionic currents of their pacing action, as well astheir gap junctional communication with host cardiomyocytes.2. Patch-clamp experiments results suggested that the allografted mHCN4-transfectedrMSCs of experimental group survived and expressed Ifcurrents in the host heart for over4weeks, while the EGFP-transfected rMSCs of in vivo control group did not express Ifcurrents.3. The allografted mHCN4-transfected rMSCs of experimental group expressed Ifwitha similar amplitude but with a more negative activation compared with parallelmHCN4-transfected rMSCs cultured in vitro(i) and in vitro(ii), characterized by a significantnegative shift of activation threshold potential and half-maximal activation voltage, andaccompanied by a seemingly slower time course with time constants. However, there wereno significant differences of reversal potential among the three groups.4. At4weeks after cell implantation, in all of the four groups of cells, neither sodiumnor calcium currents were subsequently elicited, and APs were not evoked in any group ofcells. In addition, the transplanted MSCs exhibited a low incidence of gap junctionalcoupling with host cardiomyocytes.5. At4weeks after cell implantation, humoral and cellular immune phenomena werenot observed in the area of MSCs transplantation. |
PDF Full Text Request