Nsl17555 1.4

Correlation of the electrical and intrinsic optical signals in the N.L.V. Peixotoa,*, V.M. Fernandes de Limab, W. Hankec aMicroelectronics Laboratory, University of SaÄo Paulo-USP, Esc. Politecnica, Avenue Professor Luciano Gualberto, 158, Trav.3, bDepartment of Neuroscience, FUNREI, SaÄo JoaÄo del Rey, MG, Brazil cInstitute of Zoophysiology, Hohenheim University, Stuttgart, Germany Received 11 October 2000; received in revised form 14 December 2000; accepted 18 December 2000 This paper presents some results on the correlation between the electrophysiological and intrinsic optical signals (IOS) of spreading depression waves in chicken retinae. We ®rst show that the peak of the time derivative of the electrophy- siological wave occurs precisely when the optical signal reaches the electrode tip. Second, by comparing bath applica- tions of propranolol and glycerol it can be shown that the slow potential shift is not directly correlated to the intrinsic optical signal. Propranolol depresses the amplitude of the electrical wave, although the intrinsic optical signal continues being visible. On the other hand, we observe total absence of the IOS under glycerol, while the electrical wave is always present. Correlations of this kind are relevant for a deeper understanding of the underlying mechanisms of the spreading depression phenomenon. q 2001 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Retina; Spreading depression; Chicken retinae; Propranolol; Glycerol; Intrinsic optical signals The spreading depression (SD) phenomenon was discov- Analysis of IOS and electrophysiological pro®les was ered by A. LeaÄo in 1944 [8]. Since then, SD waves have performed by treating retinae with propranolol and glycerol.
been investigated and successfully elicited in all regions of These drugs are found to in¯uence the behavior of SD the gray matter, including the retina, which constitutes its waves in different ways [12,13], but their form of action most accessible part. Intrinsic optical signal (IOS) percept- and consequences to the retinal tissue are still subject of ibility of SD is made possible due to transient changes mainly in the extracellular osmolarity, which is a side effect Experiments are carried out on 6±12-day-old chickens of the massive motion of ions between extra and intracel- [5]. Immediately after the chick being decapitated, the lular spaces. Translocation of ions also yields an electrical eyeball is dissected, cut at the equatorial plane, and the pro®le, which is present in all retinal layers during SD vitreous humour is removed. The eyecup is immersed in a waves [2,3]. Despite the fact that they may share the trigger- Petri dish perfused at a rate of 4 ml/min with Ringer solution ing event, IOS and electrophysiological signals may be (pH 7.4), containing 0.1 M NaCl, 6 mM KCl, 1 mM MgCl2, separately accessed and individually in¯uenced, as we 1 mM NaH2PO4, 1 mM CaCl2, 30 mM NaHCO3, 30 mM glucose and 5 mM Tris. In experiments with propranolol We have simultaneously measured optical (re¯ected (0.5 mM) and glycerol (5%), drugs are added to the Ringer light) and electrophysiological signals of SD waves, and solution. Temperature is maintained at 298C (^18C). Waves their synchronism has been investigated in control experi- are elicited chemically with 0.1 M KCl (approx. 50 ml ments as well as under bath application of drugs. The begin- applied by means of a micropipette) at the eyecup border every 15 min. Electrophysiological measurements are concomitant with the peak of the time derivative of the performed using glass electrodes ®lled with 1 M KCl (10 mm tip diameter) positioned in the inner plexiform layer.
The reference electrode is a silver chloride pellet electrode, * Corresponding author. Tel: 155-11-3765-1378; fax. 155-11- positioned in the bath surrounding the eyecup. Electrical signals are low-pass ®ltered at 10 Hz, and digitally recorded.
E-mail address: nathalia@lme.usp.br (N.L.V. Peixoto).
0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved.
N.L.V. Peixoto et al. / Neuroscience Letters xx (2001) xxx±xxx Optical signals are recorded in video. Electrical and optical recording systems are synchronized by an audio signal (one pulse) common to both acquisition systems. The microelec- trode becomes negative in respect to the reference electrode Concerning electrical signals, the derivative of the func- tion with respect to time is computed. At particular instants of time, selected from the electrical wave, snapshots are taken from the optical images, and a 30±40 pixels (corre- sponding to 0.20±0.27 mm of retinal tissue) wide strip oriented normally to the wavefront is cut. Its brightness pro®le is then obtained by computing the densitometry in each column [1]. Optical wave onset is considered as a 10% change in brightness as compared to the maximum attained brightness of the wave. Electrical wave duration is measured from the onset (5% as compared to the maximum) of the potential shift until the returning of the signal to the baseline as considered after the wave has propagated [11].
Amplitude of the electrical signal was found to be directly correlated to the speed of the optical wave, as measured from the IOS (that is, wavefront speed). In control experi- ments (n ˆ 51), while speed varied from 1 to 5 mm/min, amplitudes varied from 10 to 25 mV, showing a correlation coef®cient of 0.8. Moreover, derivative and amplitude have Fig. 2. Strips taken from the optical wave from Fig. 1, moving shown a correlation coef®cient of 0.85. The mean value of downward. Scale bar: 0.5 mm (75 pixels). Numbers in seconds.
the amplitude was found to be 19 ^ 1 mV (P , 0:05) and the mean speed was 3.4 ^ 0.2 mm/min (P , 0:05). Both of drugs. Fig. 3 shows brightness pro®les of two waves (one measures agree with already known values, namely, 20 of them is the wave from Fig. 2), both evaluated at the mV and 3 mm/min [10]. We found the peak of the derivative instant of time where the derivative reaches its maximum to be 17 ^ 8 mV/s (P , 0:05), occurring always before the value. In addition, brightness pro®les taken at 0 s are presented in order to exhibit the electrode tip position rela- Figs. 1 and 2 show electrical and optical concomitants tive to the waves. Note that background brightness is from a wave with bath application of propranolol. Time 0 is distorted by the pipette. It is clear from this ®gure that the arbitrarily chosen, before the wave onset. The optical wave spreads in the direction of the electrode tip, which is reached at 15.7 s. This is exactly the instant of time of the derivative peak, as taken from the electrical wave. This fact is noticed in all analyzed waves, either in controls or under the effect Fig. 3. IOS brightness pro®les (a) of a control spreading depres- Fig. 1. Slow potential shift of a wave under the effect of propra- sion wave and (b) of a wave with bath perfusion of propranolol.
nolol (amplitude 2 mV). Numbers in seconds (see Fig. 2). Polar- Pipette position is indicated on the picture. Both waves are moving to the right. Numbers near the curves in seconds.
N.L.V. Peixoto et al. / Neuroscience Letters xx (2001) xxx±xxx beginning of brightness change coincides with the pipette obtain two waves under the effect of propranolol. Never- position. This analysis has been performed in 51 control theless, the recovery of the retinal tissue is usually obtained waves and 30 waves under the effect of drugs. With a preci- by perfusing them with standard Ringer. Status of the retina sion of 0.05 s and 0.02 mm as set by the experimental is then judged from the optical composition of its back- apparatus, IOS onset happens simultaneously to the electri- Propranolol is commonly considered to be an anti- The optical wave recovers faster under the effect of migraine drug, besides many other possible uses. It is also propranolol. In Fig. 3a, a pro®le taken from a control known as a non-selective beta-adrenergic receptor-blocking wave at 11.1 s occupies most of the retinal area (in this agent [7]. Although its mechanisms of action are not well case, approx. 300 pixels, or 2 mm). On the other hand, established [4], due to the fact that its effect on the electrical under perfusion with propranolol, even during the same pro®le is much more pronounced than on the IOS, we experiment, the complete optical pro®le can be visualized conjecture that the main in¯uence of propranolol occurs at inside the screen area, as shown by Fig. 3b at 0 s and at 15.7 the early phase of ionic translocation, that is, during intense s. Moreover, the end of the electrical wave is, in this case, concomitant with the restoration of the background bright- Changes of IOS associated with neural stimulation have usually been attributed to cell volume and extracellular Under the effect of propranolol, amplitude of the electro- space alterations [9]. Because of its high viscosity, glycerol physiological signal is depressed to 48% and the derivative obstructs the water movement between the intra and extra- peak to 38% (n ˆ 10) of the control waves, whereas the cellular media, and we conjecture that during the occurrence speed is dampened to 83%. Inside the same group of of the SD this is the main cause for the depression of the waves (i.e., those under the effect of the drug) the correla- IOS. Also the higher osmolality of the perfusion solution tion coef®cient between amplitude and derivative is 0.96 can be an explanation for that, and for the form alterations (n ˆ 10). There is no alteration on the electrophysiological observed on the electrical pro®les. We would like to point waveform by comparing waves in the same experiment.
out that other optical waves can be devised [6]. Because the Bath application of glycerol completely depresses the electrical concomitant is still present, one can show the IOS. Although the electrical signal remains, speed cannot existence of ionic reorganization during wave propagation.
be measured by the previously described method. In roughly Our results show that both propranolol and glycerol half of the 22 waves analyzed in our experiments, the elec- decreased the excitability of the retinal tissue. Amplitudes trical pro®le showed an alteration in form, which is the case of the electrophysiological concomitant were lowered to 45 in Fig. 4 (notice the plateau and the second peak of the and 48% in the case of glycerol and propranolol, respec- wave). In these pro®les we consider the ®rst peak with the tively. Derivative peak dropped to 67 and 38%, respectively.
purpose of evaluating amplitude. Amplitude is lowered to Based on the electrical waveforms in control waves and 45% (n ˆ 11) of the control values, while the derivative under drug application we hypothesize that the processes peak drops to 67%. Duration is lowered to 88% (n ˆ 11) taking place at the cellular level (ionic changes and depo- of the control waves, despite the presence of the two peaks.
larization) do happen on a much faster time scale for the Retinae are very sensitive to the application of proprano- spreading of the excitation than for the reorganization of the lol, which is not the case with some other drugs [12]. This ionic composition of the extra and intracellular spaces.
implies that in the same experiment it is very dif®cult to Concomitant recording of optical and electrophysiologi- cal signals during retinal SD waves suggests the existence of separated physical processes underlying macroscopic vari- ables. Field potential appears to be an important modulator of propagation velocity. By contrast, the IOS seems to be the result of volume alterations at the cellular level.
Concerning form alteration in the case of the electrophy- siological signal with glycerol, such an alteration has not been observed in any of the control waves, or in the waves under the effect of propranolol. It remains to be investigated whether it can be a function of concentration, of the relative alteration in osmolality or of other experimental parameters such as temperature, perfusion rate or composition of the Simultaneous recording of electrophysiological and opti- cal parameters of retinal SD waves permits a demonstration of the spatio-temporal relationship between the part of the Fig. 4. Electrophysiological pro®les of a control wave (amplitude 24 mV) and of a wave under the effect of glycerol (amplitude 10 ionic ¯ux responsible for the electrical concomitant of SD waves and the intrinsic optical changes associated with N.L.V. Peixoto et al. / Neuroscience Letters xx (2001) xxx±xxx them. The electrode becomes sensitive to the electrical ®eld [5] Fernandes de Lima, V.M., Scheller, D., Tegtmeier, F., Hanke, 3±4 s before the region is invaded by the optical changes. At W. and Schlue, W.R., Self-sustained spreading depressions the derivative peak of the electrical wave the IOS onset is at in the chicken retina and short-term neuronal-glial interac- tions within the gray matter neuropil, Brain Res, 614 (1993) the pipette tip. This is a con®rmation of the customarily accepted assumption that the sensitivity of the electrode [6] Fernandes de Lima, V.M. and Hanke, W., Excitation waves occurs in a spherical volume. It also suggests that the in central grey matter: the retinal spreading depression, strength of the derivative peak, that is, the coherent ionic Prog. Ret. Eye Res, 16 (1997) 657±690.
¯ux, is a relevant component of propagation mechanisms: [7] Glaser, G.H., Penry, J.K. and Woodbury, D.M., Antiepileptic Drugs, Mechanisms of Action, Advances in Neurology, Vol.
blockage of the ®eld depresses speed of propagation and lowering of the peak derivative is compatible with smaller [8] LeaÄo, A.A.P., Spreading depression of activity in cerebral amplitudes. As we have shown, the three above-mentioned cortex, J. Neurophysiol., 4 (1944) 359±390.
parameters are not independent, but the functional relation- [9] MacVicar, B.A. and Hochman, D., Imaging of synaptically ship among them remains an open question.
evoked IOSs in hippocampal slices, J. Neurosci., 11 (1991) [10] Martins-Ferreira, H., Spreading depression in the chicken [1] Brand, S., Dahlem, M.A., de Lima, V.M.F., MuÈller, S.C. and retina, In Ookawa, T. (Ed.), The Brain and Behavior of the Hanke, W., Dispersion relation of spreading depression Fowl, Japan Scienti®c Societies Press, Tokyo, 1983, pp.
waves in chicken retina, Int. J. Bifurcation Chaos, 7 (1997) [11] Peixoto, N.L.V., Fernandes de Lima, V.M. and Hanke, W., [2] BuresÏ, J., BuresÏovaÂ, O. and KrivaÂnek, J., Meaning and Spreading depression: investigating this complex system, signi®cance of LeaÄo's spreading depression, Acad. Bras.
Proc. IEEE Comp. Soc., (1997) 147±151.
[12] Traynelis, S.F. and Dingledine, R., Role of extracellular [3] doCarmo, R.J. and Martins-Ferreira, H., Spreading depres- space in hyperosmotic suppression of potassium-induced sion of LeaÄo probed with ion-selective microelectrodes in electrographic seizures, J. Neurophysiol., 61 (1989) 927± isolated chick retina, Acad. Bras. CieÃncias, 56 (1984) 401± [13] Wiedemann, M., Fernandes de Lima, V.M. and Hanke, W., [4] Duckrow, R.B., Regional cerebral blood ¯ow during spread- Effects of antimigraine drugs on retinal spreading depres- ing cortical depression in conscious rats, J. Cereb. Blood sion, Nan. Nyn. Schmied. Arch. Pharm., 353 (1996) 552±556.

Source: http://neural.bioengineering.gmu.edu/subpages/pubs/nsl01.pdf