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 ...
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
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-
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
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 (
