review Time dependence of electric field effects on cell membranes. A review for a critical selection of pulse duration for therapeutical applications Justin Teissie 1'2, Jean Michel Escoffre1,2, Marie Pierre Rols1,2, Muriel Golzio1,2 1 CNRS; IPBS (Institut de Pharmacologie et de Biologie Structurale); 205 route de Narbonne, F-31077 Toulouse, France; 2 Universite de Toulouse; UPS; IPBS; F-31077 Toulouse, France Background. Electropulsation is one of the non-viral methods successfully used to transfer drugs and genes into living cells in vitro as in vivo. This approach shows promise in field of gene and cellular therapies. This presentation first describes the temporal factors controlling electropermeabilization to small molecules (< 4kDa) and then the processes supporting DNA transfer in vitro. The description of in vitro events brings our attention on the processes occurring before (s), during (ms) and after electropulsation (ms to hours) of DNA and cells. They all appear to be multistep events with well defined kinetics. They cannot be described as just punching holes in a lipid matrix in a two states process. Conclusions. The faster events (may be starting on the ns time scale) appear to be under the control of the external field while the slower ones are linked to the cell metabolism. Investigating the associated collective molecular reorganization by fast kinetics methods and molecular dynamics simulation will help in their safe developments for the in vivo processes and their present and potential clinical applications. Key words: electropulsation; electropermeabilization; electrotransfection; electroporation Introduction The application of electric field pulses to cells leads to the transient permeabilization of the membrane (electropermeabilization).1 This phenomenon brings new properties to the cell membrane: it becomes permeabi-lized, fusogenic and exogenous membrane Received 27 August 2008 Accepted 13 October 2008 Correspondence to: Dr. Justin Teissie, IPBS Universite de Toulouse UMR 5089 CNRS, 205 route de Narbonne, 31077 Toulouse, France. Phonel: +33(0)5 61 17 58 12; Fax: +33(0)5 61 17 59 94; Email: justin.teissie@ipbs.fr proteins can be inserted. It has been used to introduce a large variety of molecules into many different cells in vitro.2,3 Clinical applications of the electropermeabilization are now under development as a results of the EU Cliniporator and Esope programs. A local antitumoral drug delivery to patients (a method called electrochemotherapy) is under clinical trial.4-8 Transdermal drug delivery is obtained in vivo.9 More recently, electropermeabilization has been also used to transfer DNA in vivo, into the skin, liver, melanoma and skeletal muscle cells.10-15 It has the main advantages of being easy to use, fast, reproducible and safe. While during 30 years due to technological limits, pulse duration was always larger than 1 microsecond, the recent availability of high voltage (tens of kV) nanosecond long pulse generators opens the way to a new approach. Very fast perturbations under strong fields are induced in the membrane organization.16,17 A new field of development is now present for electropermeabi-lization and promising results for clinical applications were reported. One of the limiting problems remains that very few is known on the physicochem-ical mechanisms supporting the reorganisation of the cell membrane. The molecular target of the field effect remains unclear. The present review focuses on the critical role played by the pulse duration in the electropermeabilization to small molecules (< 4kDa) and on its support to the processes associated to DNA transfer in vitro. Pulse durations are easy to adjust for an optimization of the clinical target: electrochemotherapy, irreversible electropermeabilization or gene therapy as suggested as a final conclusion. Electropermeabilization Theory of membrane potential difference modulation. An external electric field modulates the membrane potential difference.18 From the physical point of view, a cell can be described as a spherical capacitor which is charged by the external electrical field. The transmembrane potential difference induced by the electric field, AYi is a complex function g(X) of the specific conductivities of the membrane (Xm), the pulsing buffer (Xout) and the cytoplasm (Xcyt), the membrane thickness and the cell size. Thus,1 in which 0 designates the angle between the direction of the normal to the membrane at the considered point on the cell surface and the field direction, E the field intensity, r the radius of the cell and f, which is a shape factor (a cell being a spheroid). Therefore, AYi is not uniform on the cell surface. It is maximum at the positions of the cell facing the electrodes. These physical predictions were checked experimentally by videomi-croscopy by using the potential difference sensitive fluorescent probes.19-21 The pulse duration plays a critical role when shorter than the capacitive loading time of the membrane. In the previous part of the paper, it was considered that the pulse was long enough to bring the potential steady state value. The loading time Tload brings a limit in this description.1 AYi = f. g (X). r. E.cos0 (1 -exp (- t / x^ ). [2] Assumming that the membrane is a true dielectric with no electric leak, the loading time. Tload, is given by Tload = rCm (1/2Xout + 1/Xcyt) [3] AYi = f. g (X). r. E.cos0 [1] Cm is the membrane capacitance, Xout and Xcyt, respectively, the conductance of the external buffer and of the cytoplasm. Tload is longer for larger cells in a heterogeneous population. Longer pulses are needed to reach the asymptotic electrically induced transmembrane voltage value (Eq. 1). A key assumption in this physical description is that the electric pulse is a sharp square wave.22 This description is under the assumption that the cell is a sphere. A more complex description is needed for spheroidal cells and their orientation relative to the field has to be taken into account.23,24 The membrane leakiness affects the loading time of the membrane when the field is applied.1 Its physical definition is given in 26 by: t= r Cm (kcyt+ 2X0Ut) / (2 XcytX0Ut + rXm (kcyt+ As Xm is dependent on the membrane leakiness, the loading time of the membrane will decrease with an increase in the membrane leakiness. The pulse duration plays a more critical role in such a case. But under physiological conditions, where Xout is larger than 10 mS/cm, as Xcyt is about 4 mS/cm, xload is always of the order of 1 |is for mammalian cells Critical parameters affecting electropermeabilization Effects of the electric field parameters. When the resulting transmembrane potential difference AY(i.e. the sum between the resting value of cell membrane AYo and the electroinduced value AYi) reaches threshold values close to 250 mV, membranes become permeable.25-26 Permeabilization is controlled by the field strength. Field intensity larger than a critical value (Ep) must be applied to the cell suspension. From Eq. [1], permeabilization is first obtained for 0 close to 0 or n. Ep is such that: AYperm = f g (X) r Ep [5] Parts of the cell surface facing the electrodes are affected. The extent of the per-meabilized surface of a spherical cell, Aperm, is given by: Aperm = Atot (1 - Ep /E)/2 [6] where Atot is the cell surface and E is the applied field intensity. Increasing the field strength (decreasing Ep/E) will increase the part of the cell surface, which is brought to the electropermeabilized state. This critical value of the transmembrane potential will be reached after a longer delay for the edges of the cap due to the loading time. But this 2 Xout)/d) [4] delay remains always in the |s time scale. This will affect the mechanism of elec-tropermeabilization only for a very short pulse duration. These theoretical predictions were assayed on cell suspension by measuring the leakage of metabolites (ATP)27 or observed at the single cell level by digitised fluorescence microscopy.28,29 The experimental results are in agreement with the predictions. The field strength must be larger than the threshold value Ep to induce permeabili-zation. The permeabilized part of the cell surface is a linear function of the reciprocal of the field intensity. Permeabilization, due to structural alterations of the membrane, remained restricted to a cap on the cell surface when short lived pulses (microseconds) are applied. The area affected by the electric field depends also on the shape (spheroid) and on the orientation of the cell with the electric field lines.24 If a train of 10 pulses is applied at a frequency of 1 Hz, it is observed that long pulses (more than 1 ms) slightly larger than Ep bring a permeabiliza-tion on two caps on the cell surface, each facing one electrode. Experimental results obtained either by monitoring conductance changes on cell suspension34 or by fluorescence observation at the single cell level microscopy28,29 shows that the local level of permeabiliza-tion is strongly controlled by the pulse du-ration.27,28 As an electrical current is flowing, Joule heating is taking place. The temperature of the sample increases as a linear function of the pulse duration and of the square of the field intensity. In vitro, this deleterious by-effect is controlled by using a low ionic content pulsing buffer to deliver a limited amount of energy. This of course cannot be controlled by that means in vivo but the tissue can be considered as a heat sink. Sieving of electropermeabilization Electropermeabilization allows a post-pulse free-like diffusion of small molecules (up to 4 kDa) whatever their chemical nature. There is a size limit for permeabilization and the process for macromolecules is described in the second part of the text. Polar small compounds cross easily the elec-tropermeabilized membrane. But the most important feature is that this membrane organisation is long-lived in cells. Diffusion is observed during the seconds and minutes following the ms pulse. Most of the exchange takes place after the pulse.28,29 Resealing of the membrane defects and of the induced permeabilization is a first order process, which appears to be controlled by protein reorganisation. For a given cell, the resealing time (reciprocal of k) is a function of the pulse duration but not of the field intensity as checked by digitised videomicroscopy.27 A precise analysis showes that several resealing processes are acting, two are very fast (ms,