Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Transcranial direct-current stimulation wikipedia , lookup
Psychophysics wikipedia , lookup
Sensory substitution wikipedia , lookup
Neurostimulation wikipedia , lookup
Total Annihilation wikipedia , lookup
Circumventricular organs wikipedia , lookup
Feature detection (nervous system) wikipedia , lookup
Stimulus (physiology) wikipedia , lookup
Perception of infrasound wikipedia , lookup
Sensory Receptors in the Anterior Uvea of the Cat's Eye An In Vitro Study Gerard M. Mintenig,* Maria V. Sdnchez-Vives,^ Carmen Martin,* Arcadi Gual* and Carlos Belmonte\ Purpose. To identify electrophysiologically the functional types of sensory fibers innervating the iris and the ciliary body of the cat's eye. Methods. The uveal tract tract of cat's eye was excised and placed in a superfusion chamber. Recordings were made from single afferent units of ciliary nerve branches responding to mechanical stimulation of the iridal surface, the ciliary body, and the choroid with a nylon filament or a glass rod. Chemical sensitivity was explored by applying acetic acid, hypertonic NaCl, and bradykinin. Warm (60°C) and cold (4°C) saline and a servocontrolled thermode were used for thermal stimulation. Results. Thirty per cent of the studied population of sensory units (n = 95) were spontaneously active when the recording was started. Approximately 30% of the fibers conducted in the lowest range of the A-delta group; the remaining 70% were C fibers. Sustained mechanical stimulation of the receptive area elicited a tonic response in approximately 60% of the units, and a phasic response in the remaining 40%. Exposure of the receptive field of mechanosensitive fibers to 600 mM NaCl evoked a long-lasting discharge in 50% of the units; application of 1 to 10 mM acetic acid elicited a short discharge in 30% of the fibers, often followed by inactivation. Bradykinin (1 to 100 fjM) produced a long-lasting response in almost 50% of the units. Warming the receptive field recruited 20% of the explored units, whereas 17% were activated by low temperature. Conclusions. Two main functional types of sensory fibers innervating the iris and the ciliary body were distinguished: (1) mechanoreceptors, corresponding to afferent units sensitive only to mechanical stimuli were generally silent at rest, had relatively higher force thresholds, and discharged phasically in response to long-lasting mechanical stimulation; (2) polymodal nociceptors, which were activated by mechanical as well as by chemical and/or thermal stimuli, usually displayed spontaneous activity, had lower force thresholds, and fired tonically upon sustained mechanical stimulation. Invest Ophthalmol Vis Sci. 1995; 36:1615-1624. A he eye receives its afferent supply from primary sensory neurons located in the trigeminal ganglion.1"3 Sensory nerve fibers innervate the ocular surface as well as various intraocular structures, including the anterior uvea (iris and ciliary body).4"8 The functional From * Uiboratori de Neurofisiologia i Biomembranes, Departamento de dearies FisuMgiques llumanes i de. la Nutririd, Universitat de Barcelona, Barcelona, Spain, and ~f Institute de Neurorienrias and Departamenlo de Fisiologia, Universidad de Alicante, Alicante, Spain Supported by grants ONCE-1989 (Spain), 94/1180 from FISSS, Ministerio de. Sanidad (Spain), and PM90-0113 and SAF93-0267 from the Comisi&n National de Cienria y Tecnologia, Direction General de Investigation Cienlifica y Tecnica (Spain). GMM was the reripient of a postgraduate fellowship from FISSS. Submitted for publication September 30, 1994; revised January 9, 1995; accepted March 13, 1995. /Proprietary interest category: N. Reprint requests: Gerard M. Mintenig, Departnment de Cienries Mediques Basiques, Faaillat de Meditina, Universitat de Lleida, Avenida Rovira Roure 44, 25198 Lleida, Catalunya, Spain. Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8 Copyright © Association for Research in Vision and Ophthalmology Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 properties of corneal and scleral sensory units are known in some detail.9"13 In contrast, information about the types of sensory receptors present in the iris and ciliary body is sparse. Neural responses evoked by mechanical stimulation of anterior uveal structures have been reported occasionally,14"16 but a categorization of the functional types of afferent units innervating these ocular tissues is still lacking. This is mainly because of the inaccessibility of intraocular structures to direct experimental manipulation. Nonetheless, detailed knowledge of the innervation of the anterior uvea is acquiring increasing clinical relevance. Implantation of intraocular lenses during cataract surgery involves manipulation of uveal structures and excitation of sensory nerves. Uveal nerves contain a variety of neuropeptides that are released during noxious 1615 1616 Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8 stimulation of the anterior segment and contribute to local inflammatory reactions (neurogenic inflammation).17 In this article, we provide electrophysiological information on the types of sensory afferent fibers innervating the iris and ciliary body of the cat obtained in an in vitro preparation of the anterior uvea. Preliminary results have been reported elsewhere.1819 METHODS Eyes from 39 adult cats of both sexes were used. All procedures were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Animals were killed with an intraperitoneal injection of 100 mg'kg" 1 of sodium pentobarbitone. Enucleation of the eye was performed when the corneal reflex was abolished and before respiratory arrest occurred. Of B Surgical Procedure Di$section of the uveal tract was carried out under a binocular microscope. The eye was placed in a chamber containing cold (4°C to 10°C) physiological saline solution (for composition, see next paragraph), and conjunctival and muscular debris were removed. The posterior hemisphere of the eye globe was divided into quadrant flaps by two perpendicular incisions that intersected at the optic disk. After removing the vitreous, retina, and lens, the choroid was carefully detached from the sclera with a cotton web soaked in saline, starting at the posterior vertex of each flap an progressing anteriorly until the iridocorneal angle was reached. This procedure exposed the ciliary nerves running in the suprachoroidal space. Nerve trunks were sectioned near the posterior pole, close to their point of entry into the sclera, and were dissected thoroughly from their connective sheath. When the dissection of all four choroid flaps and of the ciliary nerves was complete, the iridal root could be neatly cleaved off its scleral insertion. In Vitro Preparation The uveal tract with the ciliary nerves was transferred to a perspex chamber, consisting of a central bathing compartment of approximately 10 ml volume, connected to two small lateral pits, one at each side, for continuous inflow and outflow of the bathing solution (see Fig. 1). Flow was adjusted to a value of 1 to 3 ml • minute"1. The solution was kept at a constant temperature of 35°C ± 1°C by a feedback thermostatic device. The uvea, with the anterior surface up, was secured with pins to the bottom of the bathing compartment, coated with Sylgard (Dow Corning, MA). Tissue was homogeneously distended to attain the dimensions of the iris in situ. The composition of the physiological solution used to superfuse the prepara- Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 FIGURE 1. Schematic drawing showing a cross-section (A) and top view (B) of the recording chamber. 1 = perfusion inlet; 2 = dissected uveal tract in the central compartment; 3 = perfusion outlet (suction); 4 = recording amplifier. The shaded area in A is the mineral oil layer. tion was (inraM):NaCl 140, KC1 4.6, MgCl21.1, CaCl2 2.2, HEPES 10, glucose 5.6; the solution was adjusted at pH 7.4 and bubbled with oxygen. For nerve recording, the solution in the bathing compartment was covered with mineral oil. No changes in the volume or morphologic appearance of tissues were observed during the course of experimentation (up to 8 hours). Electrophysiological Recording Neural activity was recorded with an AC-coupled differential amplifier by placing nerve filaments on a platinum electrode kept within the mineral oil layer, whereas the reference electrode was immersed in saline. The output of the amplifier was filtered (0.3 to 2 kHz bandpass), displayed in an oscilloscope, and fed to a loudspeaker and a digital audiotape recorder. Nerves exhibiting activity were split longitudinally in successive steps until a single unit could be identified. A Cochet-Bonnet aesthesiometer20 with a no. 12 nylon filament (0.12-mm diameter, 0.0113-mm2 tip surface) was used to locate the receptive fields and to measure mechanical thresholds. To quantify mechanical responses, indentation pulses (duration 0.5 to 30 seconds) were applied with a round-tipped glass rod (0.64-mm diameter), mounted on a custom-built moving coil transducer driven by a pulse generator. Repetitive stimulation consisted of trains of 25 suprathreshold indentation pulses, 0.5-second duration at 0.2 Hz or 5-second duration at 0.1 Hz. Thermal stimuli consisted of a 1-ml bolus of physi- 1617 Sensory Receptors in the Anterior Uvea ological solution at 60°C or 4°C, rapidly applied to the receptive field through a thin catheter. Local temperatures in the receptive field were monitored with a fine thermistor probe gently applied on the iridal surface. Using this method, hot stimuli varied between 45°C to 55°C and cold stimuli between 4°C to 15°C. Other fibers were thermally stimulated with a flat-ended brass rod (3 mm2) attached to a Peltier cell whose temperature could be adjusted and monitored with a thermostatic regulator. Chemical sensitivity was tested by applying 0.5 to 1 ml of a 1- to 10-mM acetic acid solution or of a 600mM NaCl solution through a thin catheter placed in the vicinity of the receptive field. In a separate set of experiments, bradykinin (BK; Peninsula Laboratory Europe, Belmont, CA) was added to the perfusion solution at a final concentration of 1 to 100 ^M; bovine serum albumin (0.05% wt/vol) was used to prevent peptide adsorption to the chamber walls. Conduction velocities were calculated from the delay of action potentials evoked by suprathreshold electrical shocks (0.1 to 0.5 msec, 5 to 50 mA) applied with a pair of silver electrodes to the receptive field. The conduction distance was defined as the sum of the distance from the recording site to the point of entrance of the nerve in the tissue plus the radial distance across the iris to the center of the receptive field, and it varied from 5 to 15 mm, depending on the location of the receptive field and the length of the nerve filament. Data Analysis Recorded neural electrical activity, voltage pulses driving the electromechanic transducer, and temperature signals were replayed from tape and fed through a data acquisition interface (CED 1401; Cambridge Electronic Design, Cambridge, UK) to a computer running a software package for electrophysiological data acquisition and analysis (CED Spike2, Cambridge Electronic Design). Chi-square analysis of the distribution of qualitative variables within different groups offiberswas performed, and the Yates correction for small samples was applied when appropriate. Data are expressed as mean ± SEM or as percentages. Comparison of means was done by Student's /-test for unpaired data. Probability values lower than 0.05 were considered statistically significant. RESULTS General Single-unit recordings from 95 fibers were used for this study. Fibers responded to mechanical stimulation of the anterior surface of the iris, the ciliary body, or Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 No.of units 30 -I 25 20 15 • 10 •O2 No. of units .- 4 0 6 10 14 18 22 26 30 34 38 Receptive field (mm2) l 30 • 20 • 10 • 0 B Receptive field location 2. Size and location of uveal receptive fields. (A) Histogram showing the distribution of receptive field sizes. (B) Incidence of receptive field locations shown in the inset. Based on their locations, receptive fields were classified as (a) irido-pupillary, (b) iridal, (c) irido-ciliary, (d) ciliary, (e) cilio-choroidal, and (f) choroidal. FIGURE the choroid. Some fibers responded to areas spanning two of these structures, although most units had pure iridal fields (Fig. 2B). All fibers were considered a single group, unless differences in responsiveness associated with location of the receptive area were noticed. Receptive fields were mapped in 91 fibers using a suprathreshold value of the Cochet-Bonet aesthesiometer or a round-tipped glass rod and were generally round or oval. In twenty-four (26%) fibers, they were discontinuous, i.e., formed by two or more noncontiguous regions. A rough estimation of the surface area of the receptive field was made by multiplying its two main axes. Values varied from 1 mm2 to 36 mm2 (mean = 8 ± 1 mm2, n = 91; Fig. 2A). When the receptive area was explored with the aesthesiometer, it could be resolved into minute, discrete, sensitive points. Spontaneous activity before manipulation of the receptive field was present in 30% of the fibers. Background firing was usually irregular and continuous, with frequencies ranging from 0.5 to 10 impulses/ 1618 Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8 second '. Occasionally, some fibers fired in a bursting mode, with silent periods of variable duration. Several units that were initially silent developed ongoing activity after repeated mechanical or chemical stimulation. Miosis also appeared in some cases after repeated noxious stimulation. To investigate whether manipulations in the course of the experiment increased the incidence of spontaneous activity, we correlated the presence of spontaneous firing with the order in which the fiber was studied during the experiment. The percentages of first, second, and third units explored which displayed ongoing activity (35%, 38%, and 20% for the first, second, and third fibers, respectively) were not statistically different. Conduction velocities were calculated in 58 units (see Methods). The majority of fibers (69%) had conduction velocities under 2.0 m • second" 1 (mean = 1.0 ± 0.4 m* second" 1 , n = 40), whereas values for the remaining fibers were equal to or lower than 5 m • second" 1 (mean = 2.9 ± 1.0, n = 18). Often, fibers displaying spontaneous or mechanically evoked activity could not be recruited by electrical stimulation of the receptive area using current values up to 50 mA. However, it cannot be ruled out that in some of these cases the evoked action potentials were obscured by the large stimulus artifact produced by field stimulation. 100 % Fibres n 50 15 • % Fibres 13 0.1 0.2 1.0 2 Force threshold (mN) 11 9 7 • 5 3 1 • \\ \T, O 0- 0- 0- 0- 0' 0- 3 0 & til oft Force threshold (mN) FIGURE 3. Distribution of uveal fibers by mechanical thresholds. Force magnitudes correspond to the values given by the aesthesiometer's length scale, (inset) Cumulative frequency distribution of mechanical threshold values; mean and median values are indicated by short and long arrows, respectively. tained mechanical stimulation were explored, 46 were slowly adapting (tonic), and 30 were rapidly adapting (phasic). Both types of units were found in approximately the same proportion in the iris, the ciliary body, and the choroid. Figure 4A illustrates the response of phasic fibers to suprathreshold indentation. Increasing the amplitude or the duration of the stimulus did Mechanical Response not augment the number of evoked impulses. The Mechanosensitive units were activated by gently touchresponse of tonic fibers (Fig. 4B) was composed typiing the uveal surface with a nylon filament or a glass cally by an irregular discharge that persisted for the rod. In some cases, fibers innervating the iris could duration of the pulse and was sometimes followed by be recruited by pulling the pupillary border. a low frequency after-discharge. A deceleration of the Force thresholds were distributed across the full of firing was often observed while the stimulating rate range of the aesthesiometer (5 to 60 mm, correspondpulse was on, reflecting an adaptation of the response ing to applied forces of 0.11 mN to 1.96 mN), with to long stimuli. When trains of suprathreshold indenan average value of 0.80 ± 0.08 mN (n = 85). Approxitation pulses were applied (see Methods), fatigue was mately 15% of the units responded to low-intensity in both groups of fibers. In phasic units, faevident stimulation (<0.2 mN), whereas the majority of units tigue was characterized by an increase in the number were activated only by the higher force values of the as well as by a progressive increase of response failures, aesthesiometer (Fig. 3). Threshold was lowest in the of the first spike (Fig. 4A). Tonic units in the latency center of the receptive field. No significant differences a decrease in the number of impulses evoked showed were found in the regional distribution of mean force per stimulus, along with an increase in the latency of thresholds from the pupillary margin to the choroidal the first spike (Fig. 4B). area or in mean threshold of the fibers in the course Mechanical threshold values also were different of the experiment. in phasic and tonic fibers. By dividing the length scale The time course of the responses to mechanical of the aesthesiometer into four equal intervals, we stimulation was studied by applying a sustained suprathreshold indentation (0.5 to 30 seconds) to the re- obtained the distribution of threshold magnitudes for phasic and tonic units shown in Figure 5. Threshold ceptive field. Two types of units were distinguished, values of tonic units were homogeneously distributed according to the duration of the impulse discharge: throughout the scale range, whereas phasic fibers rapidly adapting (phasic) units that responded to intended to accumulate at high threshold values. The dentation pulses with a single spike or a short burst threshold distributions of both groups were signifilasting less than 500 msec and slowly adapting (tonic) units that gave a sustained discharge during the course cantly different (P < 0.02). of the stimulus. Tonic units roughly encoded the intensity of meAmong the 76 units for which responses to mainchanical stimulation. As shown in Figure 6, sustained Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 Sensory Receptors in the Anterior Uvea 1619 v. Frey length (cm) stimuli of increasing pressure elicited impulse discharges of progressively higher frequency. 0.0 1.5 3.0 4.5 6.0 i i i i i Chemical Response Hyperosmolar NaCl. Stimulation of mechanosensitive fibers (n = 62) with hyperosmolar (600 mM) NaCl elicited a discharge of impulses in 31 of the 76 units. The general pattern of response was a train of impulses that rapidly accelerated to a peak rate, which subsequently decayed (Fig. 7A). Ongoing activity persisted for several minutes after washing, up to 30 minutes in one fiber in which no further experimental maneuvers were made. Acetic Acid. In 30% of the fibers displaying mechanosensitivity (n = 44), 10 mM acetic acid elicited a short impulse discharge, generally lasting less than 1 minute. In most cases, the acidic stimulation led to a complete inactivation of the unit (Fig. 7B). A lower concentration (1 mM), assayed in 23 additional units, elicited firing in 12 of them. The response consisted of a rapid and brief burst of impulses, often followed 100 ms -i - Latency (ms) 64 0 5 10 15 Stimulus No. 0 5 10 15 Stimulus No. B 1 s Latency (mS) 0 5 10 15 20 Stimulus No. 60 - O Tonic • Phasic 45 % Fibres 30 15 0 ^1.96 0.98 0.35 0.16 0.11 Force threshold (mN) FIGURE 5. Distribution of the force thresholds of phasic (filled dots) and tonic (open dots) uveal fibers. Fibers with thresholds greater than the upper limit or less than the lower limit of the scale are included in the leftmost or rightmost intervals, respectively. The distributions of phasic and tonic fibers were statistically different (P < 0.02). A 4 No. Impulses 3 2 1 0 75 -. 100 -, 5025 0 0 5 10 15 20 Stimulus No. FIGURE 4. Sample recordings and fatigue of the impulse response to repeated mechanical stimulation in phasic (A) and tonic (B) uveal fibers. Suprathreshold indentantion pulses were 0.5 second at 0.2 Hz for phasic units and 5 seconds at 0.1 Hz for tonic units. In both panels, the upper trace is the recording of the fiber's response to a suprathreshold indentation pulse applied during the time indicated by the horizontal bar. The lower left histogram represents the number of impulses evoked by each stimulus, and the lower right plot (filled circles) represents the latency of the first impulse elicited by each stimulating pulse. Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 by an irregular, low frequency discharge. No inactivation was observed. All fibers sensitive to acid were excited by hypertonic NaCl. Bradykinin. The effects of BK (1 to 100 (iM) were studied in 23 mechanosensitive units whose receptive fields were located in the iris. Twelve units responded to 1 or 10 /JM concentrations of the peptide. Seven of them also were sensitive to 1 mM acetic acid. In most of the fibers (n = 10), BK induced a high frequency discharge of impulses that appeared with a latency of 1 to 30 seconds and lasted for several minutes after removing BK from the bath (Fig. 7C). In the two remaining fibers, an irregular, low-frequency train of impulses developed later than 1 minute after BK application. Units initially silent often were recruited by BK. Miosis was elicited by the peptide. Repeated applications of BK induced a progressive decrement in the response (tachyphylaxis) in three units. Increasing doses of BK, tested in two additional units, elicited larger responses when times longer than 30 minutes were allowed between applications. Thermal Response Responses to heat were explored either by application of saline solution at high temperature or by using a heating probe (see Methods). Application of 1 ml of saline solution at 60°C to the receptive field was tested in 45 units exhibiting mechanosensitivity. Nine of 1620 Force (mN) Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8 1.42 0.74 0.35 0.21 0.13 0.11 r 1 v.Frey length (cm) - 6 1)1 II I! 11 ilium IBI Illi! 10 -i Impulses.s"1 20 s them responded to heat with a prolonged, irregular discharge of impulses; three of the heat-sensitive units were tested for chemical sensitivity and showed a positive response. Four more units tested with the heating probe were recruited at a threshold temperature of 39.6°C ± 1.4°C and exhibited a discharge that peaked during the ascending phase of the stimulus and tended to be inhibited when temperature approached 50°C (Fig. 8A). In two fibers, background activity persisted for several minutes, long after temperature had returned to the control value. Sensitization of the response to heat was studied in five fibers by applying two consecutive heating ramps separated by a 5- FIGURE 6. Impulse response of an iridal tonic fiber to mechanical indentations of increasing force. The receptive field of the unit was stimulated with the aesthesiometer during the periods and at the intensities indicated by the horizontal bars (upperpanel). (middle panel) Output of the window discriminator, (loioer panel) Frequency/second"1 plot of the impulse discharge. minutes interval. One fiber, whose response to the first ramp was a sparse discharge with a threshold of 42°C, showed an increase in the number of impulses and higher instantaneous frequency peaks during the second stimulus. Cold saline solution (4°C) excited 8 of the 51 mechanosensitive fibers tested. Responses consisted of a short train of impulses lasting less than 30 seconds. A sustained bursting response was evoked by application of ice-cold solution on the iridal surface in one unit that exhibited spontaneous activity at rest. Controlled cooling down to 3°C of the receptive field with the thermal probe was performed in nine additional NaCI 600 mM A Impulses • B 20 lnnpulses-s~' 10 0 C BK 100 20 • Impulses-s" 1 1 0 • J 0• ^1v^VUrV^w"|^MrtJ^^ 20 s FIGURE 7. Frequency/second ' plots of the responses of three different iridal fibers to a bolus application on the receptive field (arrows) of 600 mM NaCI (A), 10 mM acetic acid (B), and 100 fiM bradykinin (C). Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 10 s FIGURE 8. Response of two different uveal fibers to heating and cooling ramps applied to the receptive area with a thermostated probe. (A) Impulse discharge evoked by gradual heating. (B) Response to cooling. For both panels: upper graph, frequency/second"1 plot; lower trace, stimulus temperature. Sensory Receptors in the Anterior Uvea 1 [ 100 -i ** 1621 1 Polymodal 1 Mechanoreceptive *•* - 1.2 mN % Fibres 75 - 1.0 50 25 0 J Sp. Active Tonic FIGURE 9. Functional characteristics of polymodal and pure mechanoreceptive uveal fibers, (left) Incidence of spontaneous activity and of tonic response to mechanical indentation in both classes of units, (right) Force thresholds of polymodal and mechanoreceptive fibers. Values are mean ± SEM of 35 polymodal and 26 mechanoreceptive fibers. *P < 0.02; **P< 0.005; ***P< 0.001. mechanosensitive units. In three of them, discharges were elicited at a mean threshold of 15.1°C ± 1.9°C. Activity persisted until temperature returned to control values. A sample record of the response to cooling in one of these fibers is shown in Figure 8B. Differences Between Pure Mechanosensitive Units and Polymodal Units Based on the presence of chemical or thermal sensitivity, or both, uveal units were classified as pure mechanosensitive (responding only to mechanical force) and polymodal units, which were also activated by chemical or thermal stimulation, or both. Figure 9 compares the main functional characteristics of both classes of units. The incidence of spontaneous activity at the beginning of the experiment was significantly higher in polymodal fibers (46% in polymodal, n = 35, versus 9% in mechanoreceptive, n = 34; P < 0.005). Additionally, polymodal fibers responded more frequently to long indentation pulses with a tonic discharge than to mechanoreceptive units (83% in polymodal, n = 30, versus 32% in mechanoreceptive, n = 25; P < 0.001). Furthermore, mean mechanical threshold was significantly lower in polymodal units in comparison with pure mechanosensitive units (0.67 ±0.11 mN versus 1.05 ± 0.14 mN for polymodal and mechanoreceptive units, respectively; P < 0.02). Other Types of Units Occasionally, units that started to fire by application of acid or hypertonic NaCl were observed while recording from a filament containing a mechanosensitive unit of different amplitude or shape. These chemosensitive units could not be recruited by mechani- Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 cal stimulation, but no attempts were made to characterize them further. Some of the units displaying a rhythmic spontaneous activity were not affected by mechanical stimulation of the receptive area but increased their firing frequency when cold saline was applied to that area. The responsiveness of these units to more controlled thermal stimuli was not explored in detail. DISCUSSION Our results show that the anterior uvea is innervated by different functional types of sensory afferent units. In her pioneering work, Tower14 described multiunit responses in ciliary nerves of the cat that were evoked by touching the anterior surface of the iris or by pushing the lens. These early results were confirmed by single unit recordings of responses to direct stimulation of the iris or intraocular pressure elevations.41516'21 In the intact eye, however, receptive fields are difficult to access, and the functional properties of sensory fibers cannot be studied in detail. We have developed an in vitro preparation that permitted us to analyze the functional properties of uveal sensory innervation. No changes in functional characteristics or sensitivity of the studied units were detected during the recording time (up to 8 hours). Similar survival times were reported for sensory afferents in other superfused tissues.13'22"24 Morphologic reports5'6'25"27 have described myelinated and unmyelinated sensory fibers in the irisciliary body of various species. This was confirmed electrophysiologically in this work, in which a majority of fibers conducted in the Ofiber range. Conduction velocity measurements in our experimental conditions had several potential sources of error: Estimations of conduction distances did not take into account the loss of the myelin sheath and the tortuous path of fibers within the iridal parenchyma6'25'28; furthermore, stimulus artifacts were large in relation to the short conduction distances, leaving open the possibility that the evoked potential of fast fibers was obscured by the artifact. In spite of these limitations, the proportion of myelinated fibers found in our work (approximately 30%) is close to the value estimated by morphologic methods.5 The number of fibers responding to mechanical stimulation was large in the peripheral zone of the iris in comparison to the pupillary border, the ciliary body, or the choroid. This observation also agreed with morphologic data, which reported a decrease in density of sensory innervation from the root of the iris toward the pupillary margin.5 The sampling method used in this work (presence of mechanosensitivity in the studied unit) was aimed to reduce to a minimum repeated noxious stimulation of tissues. However, this procedure excludes sensory 1622 Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8 fibers responding only to other forms of energy, such as thermal receptors or mechanically insensitive nociceptors. In fact, units recruited only by chemical stimuli or by cold were occasionally encountered in this study and may indicate the existence of mechanically insensitive sensory fibers in the uvea. Taking this possibility into account, two main functional classes of uveal sensory fibers were distinguished in this study, based on the characteristics of their responses to different stimulus modalities: pure mechanoreceptive and polymodal fibers. Pure mechanoreceptive fibers, responding only to mechanical stimuli, usually did not fire spontaneously, had higher force thresholds, and predominantly phasic responses. Moreover, their mechanical threshold values were equivalent to those found in nociceptive units of other areas, such as the testis or the cornea."'23 The same is true of the firing pattern elicited by repeated mechanical stimulation. These properties make them analogous to highthreshold mechano-nociceptors of the skin and outer coats of the eye."12'29'30 The possibility remains, however, that some of the mechanosensitive phasic fibers recorded in vitro could be excited by physiological changes in pupillary diameter in vivo. In contrast, polymodal units were sensitive to irritant solutions and to heat or cold, and they showed higher incidences of spontaneous activity. In most cases, their responses to mechanical stimulation were tonic. Only a few units that responded to heat could be explored with chemicals, although all of them exhibited chemical sensitivity. Furthermore, the proportion of mechanosensitive fibers responding to heat was lower than those responding to chemicals, perhaps because, to avoid irreversible damage to the tissue, no temperatures higher than 60°C were used, and the rapid cooling may have restricted stimulation to superficial units. In spite of these limitations, the functional characteristics of our polymodal fibers closely resembled those of polymodal nociceptors of other territories.91112'31'32 As occurs with polymodal units in various tissues, irritant chemicals such as hypertonic NaCl, protons, or BK effectively stimulated polymodal uveal afferents. Sensitivity of uveal units to acidic solutions was high compared to corneal afferents,1112 and inactivation was often obtained with 10 mM acetic acid applied to the receptive field. Bradykinin is an algesic substance produced in injured tissues; it acts as a potent stimulator of polymodal nociceptivefibers.33"34Short latency responses of iridal units to BK are attributable to a direct effect of this substance on nociceptive terminals, whereas delayed discharges more likely are caused by a mechanical stimulation secondary to contraction of sphincter pupillae muscle.35 Responses to 60°C saline solution applied on the receptive field were present in approximately 20% of the units. The temperature rises obtained with this Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 method are transient; thus, the number of heat-sensitive units may be underestimated. Thermal thresholds and firing pattern under gradual heating of the receptive area were similar to those found in polymodal nociceptors of the cornea.9'12 Although responses to repeated heating were not studied in detail, sensitization36 seems to occur in some of the uveal polymodal fibers. A weak sensitivity to cooling commonly is present in polymodal nociceptors931'37'38 and was found in uveal polymodal fibers. Spontaneous activity was present in a significant proportion of uveal units. In part, this may have been caused by the stretch of tissues pinned to the bottom of the perfusion chamber, but it also may have been a consequence of sensitization of nociceptive terminals. It is well known that ongoing activity develops in a proportion of sensitized polymodal nociceptors.36'39'40 Sensitization results from the action of inflammatory mediators released by injured tissues (see ref. 41). In the iris, mechanical and thermal stimulation releases prostaglandins and neuropeptides that produce miosis.42'43 Thus, it is conceivable that trauma during tissue extraction had some sensitizing action on uveal polymodal nociceptors. In turn, stimulated nociceptors release neuropeptides, such as substance P, that enhance local inflammatory reactions.17'44 This may explain the appearance of miosis, sometimes observed in our experiments after repeated stimulation of the iris. The different uveal tissues (iris, ciliary body, choroid) appear to be equally innervated by high threshold mechano-nociceptive and polymodal nociceptive fibers. The abundance of uveal nociceptors explains the intense pain that accompanies various forms of uveitis, as well as the pain arising from the contracture of the ciliary muscle. On the other hand, new techniques in ophthalmology include photocoagulation and implantation of intraocular lenses. Frequendy, they involve manipulation and injury to uveal structures, accompanied by stimulation of uveal nociceptors. Neurogenic inflammation developed by the excitation of the uveal nociceptors described here presumably potentiates the inflammatory reaction of uveal tissues after surgical trauma or other local pathologic processes. Key Words ciliary body, iris, mechanoreceptors, nociceptors Acknowledgments The authors thank Dr. R. Gallego for critical reading of the manuscript, and they thank Mr. Simon Moya and Mr. Alfonso Perez-Vegara for technical assistance. References 1. Ardvison B. Retrograde axonal transport of horserad- Sensory Receptors in the Anterior Uvea 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. ish peroxidase from cornea to trigeminal ganglion. Ada Neuropathol (Berl). 1977;38:49-52. Morgan CW, Nadelhaft I, deGroat WC. Anatomical localization of corneal afferent cells in the trigeminal ganglion. Neurosurgery. 1978;2:252-258. Marfurt CF. Somatotopic organization of the cat trigeminal ganglion as determined by the horseradish peroxidase technique. Anat Rec. 1981;201:105-118. Lele PP, Weddell G. Sensory nerves of the cornea and cutaneous sensibility. Exp Neurol. 1959; 1:334-359. Bergmanson JPG. The ophthalmic innervation of the uvea in monkeys. Exp Eye Res. 1977; 24:225-240. Lehtosalo JI, Uusitalo H, Palkama A. Sensory supply to the anterior uvea: A light and electron microscope study. Exp Brain Res. 1984;55:562-569. Ayer-Lelievre C, Granholm ACH, Seiger A. Silver impregnation of intrinsic sensory innervation of rat iris in toto. J Anat. 1984; 138:323-335. Olson L, Ayer-Le Lievre C, Bjorklund H, et al. The innervation apparatus of the rodent iris. In: Bjorklund A, HokfeltT, Owan C, eds. Handbook of Chemical Neuroanalomy. Vol. 6. The Peripheral Nervous System. Amsterdam: Elsevier; 1988:545-597. Belmonte C, Giraldez F. Responses of cat corneal sensory receptors to mechanical and thermal stimulation. JPhysiol. 1981;321:355-368. Tanelian DL, Beuerman RW. Responses of rabbit corneal nociceptors to mechanical and thermal stimulation. Exp Neurol. 1984; 84:165-178. Belmonte C, GallarJ, Pozo MA, Rebollo I. Excitation by irritant chemical substances of sensory afferent units in the cat's cornea. JPhysiol. 1991;437:709-725. GallarJ, Pozo MA, Tuckett RP, Belmonte C. Response of sensory units with unmyelinated fibres to mechanical, thermal and chemical stimulation of the cat's cornea. JPhysiol. 1993; 468:609-622. Mclver MB, Tanelian DL. Structural and functional specialization of Ad and C fiber free nerve endings innervating rabbit corneal epithelium. J Neurosci. 1993; 13:4511-4524. Tower SS. Units for sensory reception in the cornea: With notes on nerve impulses from sclera, iris and lens. / Neurophysiol. 1940; 3:486-500. Lele PP, Grimes P. The role of neural mechanisms in the regulation of intraocular pressure in the cat. Exp Neurol. 1960; 2:199-220. Zuazo A, Ibanez J, Belmonte C. Sensory nerve responses elicited by experimental ocular hypertension. Exp Eye Res. 1986;43:759-769. Unger WG. Mediation of the ocular response to injury and irritation: Peptides versus prostaglandins. In: Bito LZ, Stjernschantz J, eds. The Ocular Effects of Prostaglandins and Other Eicosanoids. New York: Alan R. Liss; 1989:293-298. Belmonte C, Mintenig GM, Gual A. Functional properties of sensory receptors of the iris and ciliary body. ARVO Abstracts. Invest Ophthalmol Vis Sci. 1988; 29:18. Sanchez-Vives MV, Belmonte C. Effects of peptides on nervous activity of sensory fibres in the iris of the cat. ARVO Abstracts. Invest Ophthalmol Vis Sci. 1990;31:514. Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 1623 20. Cochet P, Bonnet R. L'esthesie corneene. Clin Ophthalmol. 1960; 4:1-27. 21. Perkins ES. Sensory mechanisms and intraocular pressure. Exp Eye Res. 1961;l:160-167. 22. Sharkey KA, Cervero F. An in vitro method for recording single unit afferent activity from mesenteric nerves innervating isolated segments of rat ileum. / Neurosci Meth. 1986; 16:149-156. 23. Kumazawa T, Mizumura K, Sato J. Response properties of polymodal receptors studied using in vitro testis-superior spermatic nerve preparations of dogs. J Neurophysiol. 1987; 57:702-711. 24. Reeh P. Sensory receptors in mammalian skin-nerve in vitro preparation. Prog Brain Res. 1988; 74:271-276. 25. Saari M. Organization of the myelinated nerves in the cat iris. Exp Eye Res. 1973; 17:275-279. 26. Saari M, Kiviniemi P. Wallerian degeneration of the myelinated nerves of cat iris after denervation of the ophthalmic division of the trigeminal nerve. Exp Eye Res. 1973; 17:281-287. 27. Huhtala A, Johansson G, Saari M. Myelinated nerves of guinea pig iris after denervation of the ophthalmic division of the trigeminal nerve. Ophthalmic Res. 1975; 7:354-362. 28. Lehtosalo JI, Uusitalo H, Uusitalo J, Poranen A, Palkama A. Electron microscopic identification of trigeminal nerve fibers in the ciliary body and the iris. Neurosci Lett. 1983;42:239-242. 29. Burgess PR, Perl ER. Myelinated afferent fibres responding specifically to noxious stimulation of the skin. J Physiol. 1967; 190:541-562. 30. Georgopoulos AP. Functional properties of primary afferent units probably related to pain mechanisms in primate glabrous skin. /Neurophysiol. 1976;39:71-83. 31. Bessou P, Perl ER. Response of cutaneous sensory units with unmyelinated fibres to noxious stimuli. J Neurophysiol. 1969;32:1025-1043. 32. Campbell JN, Meyer RA, LaMotte RH. Sensitization of myelinated nociceptive afferents that innervate monkey hand./Neurophysiol. 1979;42:1669-1679. 33. Beck PW, Handwerker HO. Bradykinin and serotonin effects on various types of cutaneous nervefibre.Pflu'gersArch. 1974,347:209-222. 34. Kumazawa T, Mizumura K. Chemical responses of polymodal receptors of the scrotal content of the dog. JPhysiol. 1980; 299:219-231. 35. Butler JM, Hammond BR. The effects of sensory denervation on the responses of the rabbit eye to prostaglandin El, bradykinin and substance P. BrJPharmacol. 1980; 69:495-502. 36. Fitzgerald M, Lynn B. The sensitization of high threshold mechanoreceptors with myelinated axons by repeated heating. JPhysiol. 1977; 365:549-563. 37. Iggo A. Cutaneous heat and cold receptors with slowly conducting (C) afferent fibres. Quart J Exp Physiol. 1959;44:362-370. 38. Kumazawa T, Mizumura K. Mechanical and thermal responses of polymodal receptors recorded from the superior spermatic nerve of dogs. / Physiol. 1980; 299:233-245. 39. Perl ER. Sensitization of nociceptors and its relation 1624 Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8 to sensation. In: Bonica JJ, Albe-Fessard D, eds. Advances in Pain Research and Therapy. Vol. 1. New York: Raven; 1976:17-28. 40. Schaible HG, Schmidt RF. Time course of mechanosensitivity changes in articular afferents during a developing experimental arthritis. J Neurophysiol. 1988; 60:2180-2195. 41. Handwerker HO. What peripheral mechanisms contribute to nociceptive transmission and hyperalgesia?. In: Basbaum AI, Besson JM, eds. Towards a New Pharmacotherapy of Pain. Chichester: John Wiley & Sons; 1991:5-19. Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 42. Cole DF, Unger WG. Prostaglandins as mediators for the response of the eye to trauma. Exp Eye Res. 1973; 17:357-368. 43. Zhang QS, Butler JM, Ohara K, Cole DF. Sensory neural mechanisms in contraction of the isolated sphincter pupillae: The role for substance P and the effects of sensory denervation on the response to miotics. Exp Eye Res. 1982; 35:43-54. 44. Bill A, StjernschantzJ, Mandahl A, Brodin E, Nilson G. Substance P: Release on trigeminal nerve stimulation, effects on the eye. Ada Physiol Scand. 1979; 106:371373.