Download Sensory receptors in the anterior uvea of the cat`s eye. An in

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.