Wide-band spectral tuning of heat receptors in the pit
organ of the copperhead snake (Crotalinae)
Vera Moiseenkova, Brent Bell, Massoud Motamedi, Edward Wozniak and Burgess
Christensen
Am J Physiol Regul Integr Comp Physiol 284:R598-R606, 2003. First published 7 November 2002;
doi: 10.1152/ajpregu.00024.2002
You might find this additional info useful...
This article cites 26 articles, 7 of which you can access for free at:
http://ajpregu.physiology.org/content/284/2/R598.full#ref-list-1
Updated information and services including high resolution figures, can be found at:
http://ajpregu.physiology.org/content/284/2/R598.full
Additional material and information about American Journal of Physiology - Regulatory, Integrative
and Comparative Physiology can be found at:
http://www.the-aps.org/publications/ajpregu
This information is current as of May 25, 2013.
American Journal of Physiology - Regulatory, Integrative and Comparative Physiology publishes original
investigations that illuminate normal or abnormal regulation and integration of physiological mechanisms at all
levels of biological organization, ranging from molecules to humans, including clinical investigations. It is published
12 times a year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991.
Copyright © 2003 the American Physiological Society. ISSN: 0363-6119, ESSN: 1522-1490. Visit our website at
http://www.the-aps.org/.
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
This article has been cited by 4 other HighWire-hosted articles:
http://ajpregu.physiology.org/content/284/2/R598#cited-by
Am J Physiol Regul Integr Comp Physiol 284: R598–R606, 2003.
First published November 14, 2002; 10.1152/ajpregu.00024.2002.
Wide-band spectral tuning of heat receptors in the pit
organ of the copperhead snake (Crotalinae)
VERA MOISEENKOVA,1,2 BRENT BELL,1 MASSOUD MOTAMEDI,1
EDWARD WOZNIAK,3 AND BURGESS CHRISTENSEN2
1
Center for Biomedical Engineering, The University of Texas Medical Branch, Galveston;
3
Animal Resources Center, The University of Texas Medical Branch, Galveston; and 2Department
Physiology and Biophysics, The University of Texas Medical Branch, Galveston, Texas 77555
Moiseenkova, Vera, Brent Bell, Massoud Motamedi,
Edward Wozniak, and Burgess Christensen. Wide-band
spectral tuning of heat receptors in the pit organ of the
copperhead snake (Crotalinae). Am J Physiol Regul Integr
Comp Physiol 284: R598–R606, 2003. First published November 14, 2002; 10.1152/ajpregu.00024.2002.—Receptors
located in the facial pit organ of certain species of snake
signal the presence of prey. Infrared radiation is an effective
stimulus suggesting that these receptors may be low-threshold temperature receptors. We recorded from the nerve innervating the pit organ of snakes belonging to the family of
Crotalinae while stimulating the receptive area with welldefined optical stimuli. The objective was to determine the
sensitivity of these receptors to a wide range (0.400–10.6 ␮m)
of optical stimuli to determine if a temperature-sensitive or
photosensitive protein initiated signal transduction. We
found that receptors in the pit organ exhibited a unique
broad response to a wide range of electromagnetic radiation
ranging from the near UV to the infrared. The spectral
tuning of these receptors parallels closely the absorption
spectra of water and oxyhemoglobin, the predominant chromophore in tissue. Our results support the hypothesis that
these are receptors activated by minute temperature changes
induced by direct absorption of optical radiation in the thin
pit organ membrane.
pit vipers; rattlesnakes; temperature sensation; thermosensation
THE PIT ORGAN is a receptive organ located in the head of
venomous snakes of the subfamily Crotalinae. Receptors in the pit membrane are unique structures that
enable the snake to identify nearby prey by sensing
their temperature. Crotalines, known collectively as pit
vipers, include rattlesnakes, moccasins, copperhead,
and a number of other New World vipers (20). How the
pit organ is used to sense prey in the wild has been the
subject of several studies (11, 12, 14, 24), and a fundamental question is whether the receptors behave as
photoreceptors with a photosensitive pigment or thermal receptors expressing a unique low temperaturesensitive protein. It is generally accepted that the pit
organ receptors are highly sensitive to electromagnetic
Address for reprint requests and other correspondence: B. N. Christensen, Dept. Physiology & Biophysics, Route 0437, Univ. of Texas
Medical Branch, Galveston, TX 77555 (E-mail: [email protected]).
R598
radiation in the infrared region of the spectrum, which
would be consistent with their role as heat receptors.
In fact, most studies have focused on determining the
spectral sensitivity of the receptors by recording activity from the afferent nerve fibers in the trigeminal
nerve to infrared radiation (7, 13, 14, 19).
On the basis of behavioral studies, Noble and
Schmidt (25) proposed that snakes could detect their
prey by sensing body temperature with a sensitivity of
0.2°C. Using a variety of thermal stimuli, de Cock
Buning (13) measured activity in nerve fibers over a
temperature range of 15–33°C. Consistent with this
report, electrophysiological recordings from the trigeminal nerve of the rattlesnake demonstrated that
warm objects trigger nerve impulses (7). They concluded that the pit organ receptors are specialized heat
receptors because of the agreement between the calculated temperature rise with the response threshold for
infrared stimulation. Their measurements gave a sensitivity of 0.001–0.005°C.
The extraordinary temperature sensitivity of heat
receptors in the pit organ suggests that either the
receptors function as an infrared photoreceptor (17) or
there are highly unique heat receptors in the pit that
can respond to a minute heat stimulus. A high sensitivity restricted to infrared (IR) radiation may suggest
the presence of unique photoreceptors in the pit that
can detect IR directly through a photochemical interaction. On the other hand, a broad spectral response
(400–10,600 nm) of the pit organ would be inconsistent
with the spectral sensitivity of known photosensitive
pigments.
Previous studies have neither addressed the spectral
response of thermal receptors in the pit organ in the
visible and ultraviolet region of the spectrum nor the
dependence of the receptor response on the dose of
optical radiation as a function of the wavelength and
the strength of the optical stimulus. Special heat receptors in the pit organ could function primarily as
temperature sensors that respond directly to a local
change in temperature of the pit membrane. This sugThe costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
0363-6119/03 $5.00 Copyright © 2003 the American Physiological Society
http://www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
Submitted 16 January 2002; accepted in final form 3 November 2002
R599
SNAKE PIT ORGAN HEAT RECEPTORS
MATERIALS AND METHODS
Copperhead (Agkistrodon contortrix) snakes were used in
all of the experiments reported here. Some of the copperheads were captive bred and ranged in age from 10 mo to 2
yr. The wild copperhead ages ranged from 1 to 25 yr. The
animals were obtained through reputable vendors (Glades
Herp, Myers, FL) and housed in The University of Texas
Medical Branch (UTMB) animal facilities. All snakes were
individually housed in appropriately sized glass aquariums
equipped with a water bowl, hide box, rock, and focal heat
source to allow for behavioral thermoregulation. All cages
were maintained in a centralized secure room with an
ambient temperature of 24–28°C and a 12:12-h light-dark
cycle. Routine care was provided through the UTMB Animal Resources Service. All snakes were given a 2- to 4-wk
acclimation period before being used in any experimental
procedures. All procedures involving the use of animals
were conducted in accordance with an approved National
Institutes of Health Guide for the Care and Use of Laboratory Animals and animal protocol. The lengths and
weights ranged from 60 to 130 cm and from 100 to 700 g,
respectively.
Anesthesia induction for each snake occurred in two steps.
Personnel in the UTMB animal care facility trained to handle
venomous snakes carried out all procedures necessary for
anesthetization. First, the snake was placed into an airtight
plastic container (Sterilite, Townsend, MA) with an Isoflurane USP (Abbott Laboratories, North Chicago, IL) soaked
gauze (4 ⫻ 4). The snake was then visually monitored until
no movement could be observed (usually 30–45 min). Once
fully under the influence of the gas anesthetic, the animal
was carefully taken from the plastic container and the poison
fangs were removed by cutting. Next, the animal was intubated using a 1- to 3-mm-diameter flexible tracheal tube and
the poison glands were removed. Continuous anesthesia for
the snake was maintained using a small animal anesthesia
machine. The anesthesia machine included a ventilator (Harvard Apparatus, Holliston, MA) coupled to an Isoflurane
vaporizer (Primary Medical Products, Los Angeles, CA). Animals were ventilated with 100% oxygen and 0.5–4% Isoflurane. Ventilator tidal volumes ranged from 15 to 50 ml and
were adjusted for each snake depending on the size. Respiration rates were maintained between 8 and 12 breaths/min.
AJP-Regul Integr Comp Physiol • VOL
For nerve preparation and recording, the animal was
moved to a Sylgard (Dow Corning, Midland, MI) filled Plexiglas tray. The Plexiglas tray was designed such that the
snake’s body remained at a constant level from head to tail.
Careful positioning on the tray ensured that experiments
were conducted with the animal in as close to natural state as
possible. In the tray, the distal two-thirds of the snake’s body
were on top of a warming pad. The warming pad maintained
the surface temperature between 27 and 30°C as measured
by a skin surface positioned Type T thermocouple and readout unit (Omega Engineering, Stamford, CT). The thermocouple was affixed to the skin on the dorsum of the animal
using an insulated patch with pressure-sensitive adhesive.
Heart rate was monitored using a Hewlett Packard 78353B
EKG machine (Hewlett Packard, San Francisco, CA) and
subcutaneous needle electrodes. The heart rate during experiments ranged from 45 to 60 beats/min.
The head of the snake was positioned laterally on the tray
to expose the left or right cheek area. Immobilization and
fixation of the head and upper body were accomplished using
pins pushed through the skin of the lower jaw and into the
Sylgard.
The maxillary branch of the trigeminal nerve was dissected and cut as far centrally as possible. The connective
sheath surrounding the nerve was stripped back. The whole
nerve bundle was placed on silver chloride wire bipolar electrodes and covered with mineral oil. The electrodes were
connected to a differential amplifier (Warner Instrument,
Hamden, CT) and the signal was amplified 10,000 times and
low- and high-pass filtered at 300 Hz and 10 kHz using
cascaded Bessel and Butterworth filters (8 pole). Data were
recorded directly to a PC-based acquisition system using the
software AxoScope (Axon Instruments, Union City, CA) at a
sampling rate of 100 kHz. Raw data files were imported to
pClamp version 8 software suite (Axon Instruments) and the
Clampfit program was used to convert the data to the frequency domain using a Fast Fourier Transform.
To optically stimulate the pit organ, we used six different
optical sources to produce eight independent stimuli. They
included five independent lasers, Helium-Neon (633; 1,150;
3,390 nm) (NEC, Tokyo, Japan and Research Electro-Optics,
Boulder, CO), argon-ion (488 and 514 nm) (Spectra-Physics,
Mountain View, CA), diode (1,450 nm) (OptoPower, Tucson,
AZ), carbon dioxide (10,600 nm) (Laser Industries Limited,
Tel-Aviv, Israel), and mercury light source with filter (400
nm) (EFOS, Ontario, Canada). Laser power was attenuated
using variable attenuators (II-VI Incorporated, Saxonburg,
PA and Newport, Irvine, CA) and ranged from 0.035 to 8 mW.
Optical radiation from lasers was delivered to the pit organ
through different optical fibers depending on the desired
wavelengths as follows: 400 nm, low OH fused silica (band
pass 300–2,200 nm); 488; 514; 632; 1,450; 3,390 nm; sapphire
optical fiber (band pass 500–3,200 nm) (Saphikon, Milford,
NH); 10,600 nm; silver-halide optical fiber (band pass 3,000–
15,000 nm) (Panama City Beach, FL). Each fiber was positioned such that it directly irradiated the pit organ of the
snake without contact. A mechanical shutter (Uniblitz, Rochester, NY) placed between the laser and the optical fiber was
used to vary the stimulus duration from 8 ms to 10 s. The
laser beam was never greater than the diameter of the pit
organ aperture, which is ⬃1–1.5 mm. Distance from the
optical fiber tip to snake pit organ was 3 mm or less. When
the shutter was closed, no response above background noise
was recorded from the nerve bundle.
To chemically stimulate the pit organ, capsaicin (50 ␮M in
ethanol) was applied directly to the pit organ membrane.
Alcohol alone served as control for the vehicle.
284 • FEBRUARY 2003 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
gests that a temperature-sensitive protein with unique
ability to respond to a low-temperature stimulus is
present in the pit organ that is in principle similar to
the vanilloid receptor that has been recently discovered
in mammalian tissue (10). However, the morphological
or molecular basis conferring infrared or heat sensitivity of the pit organ receptors has not been reported yet.
In this study, we investigated the properties of the
pit organ receptors using electromagnetic radiation of
discrete frequencies that span the spectrum from the
UV to the infrared. We measured nerve activity as a
function of wavelength, fluence rate, and stimulus duration in copperhead snakes and demonstrated that
the pit organ has a very broad spectral response and is
most sensitive to those wavelengths that are best absorbed by water or oxyhemoglobin, two chromophores
that are abundant in tissue of the pit organ and could
participate in photothermal interaction inducing
highly localized temperature changes in tissue during
exposure to optical radiation.
R600
SNAKE PIT ORGAN HEAT RECEPTORS
RESULTS
Fig. 1. Data acquisition and processing. A: typical raw data of whole nerve
response to 1-s stimulus from an argon-ion laser (514 nm, fluence rate was
116 mW/cm2). B: spectral analysis of
the response in A. The area under the
power spectral density (PSD) curve between 350 and 650 Hz was measured.
C: family of PSDs for different stimulus fluence rates. For data presentation only, the PSDs have been
smoothed between 100 and 1,000 Hz
using a running average algorithm.
The actual calculation of the area was
made from the nonsmoothed PSDs like
that shown in B. The area was measured under the PSD curve in the region between 350 and 650 Hz as indicated by the vertical lines on the
graph. D: dependence of the magnitude
of the area under the PSD curve on the
stimulus fluence rate for a stimulus of
1-s duration at 514 nm. The data
points are means and SD from 5 trials.
AJP-Regul Integr Comp Physiol • VOL
284 • FEBRUARY 2003 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
In these experiments, we recorded from the whole
nerve and measured the power spectral density (PSD)
of the response as representative of the nerve activity.
The area under a fixed bandwidth (350 to 650 Hz) was
calculated and arbitrarily chosen as representative of
the nerve activity to a specific stimulus. Figure 1A
shows a typical response from the whole nerve to a 1-s
application of optical radiation (514 nm) with fluence
rate of 116 mW/cm2 and the PSD for this response is
shown in Fig. 1B. Similarly, to control for background
noise, the PSD was calculated just before the stimulus
and the area between 350 and 650 Hz was measured.
This background was subtracted from the area calculated from the PSD of the response during the stimulus
application. Figure 1C shows the magnitude of the
PSDs for three different stimulus strengths for the
same laser (514 nm). Vertical lines indicate the area
under each PSD from 350 to 650 Hz illustrating where
the calculations were made. This area is representative
of the response of the system. A stimulus-response
curve is shown in Fig. 1D and shows a relatively linear
rise reaching saturation near a fluence rate of 125
mW/cm2. The advantage of this approach compared
with single-fiber recording is that there is no bias to
fiber size as activity from the whole nerve is recorded.
In addition, in a single experiment, many different
stimuli could be applied to the sensory organ with very
stable recording from the nerve. The single major potential disadvantage of whole nerve recording is that if
the fiber population is functionally heterogeneous, specific information on sensory modality and firing characteristics is lost as only the average response proper-
ties may be obtained. However, single-fiber recording
has demonstrated that the pit organ is composed of
nerve fibers with very similar firing characteristics (7).
Another disadvantage is that comparison of absolute
magnitude of response between preparations is not
possible because the size of the response depends
greatly on the particular nerve preparation and recording conditions, which varied from animal to animal.
For this reason, to compare responses from different
wavelength optical stimuli, data have been normalized
to the largest response obtained from the most effective
wavelength used in an experiment.
It is common to record activity from primary afferent
nerve fibers under general anesthesia because centrally acting anesthetics usually have little influence
on peripheral nerve. However, we found that the halogenated anesthetic used in these experiments had a
depressive effect on the pit organ nerve activity. Figure
2A shows the effects of anesthesia on the receptor
organ response threshold. All of these data were recorded from the same nerve preparation in a copperhead snake. Three wavelengths (514, 632, and 1,450
nm) were used to stimulate the pit organ during the
delivery of the isoflurane at concentrations ranging
from 0.5 to 4%. At each anesthetic concentration, each
stimulus was delivered to the pit organ in sequence.
The stimulus strength necessary to produce a measurable response in whole nerve activity was measured.
Two observations can be made from these results.
First, the sensitivity of the pit organ to the four wavelengths is different. Second, the stimulus strength necessary to produce a measurable response increased as
the anesthetic dose was increased independent of
R601
SNAKE PIT ORGAN HEAT RECEPTORS
Fig. 2. Effect of anesthesia on nerve response. A: receptor threshold
taken as the fluence rate necessary to produce a detectable response
in the nerve as a function of increased percentage of anesthesia was
measured from the same preparation for wavelengths of 514, 632,
and 1,450 nm. The order of anesthetic concentration delivered was
1.5, 1, 4, and 0.5%. There was an interval of at least 20 min between
changes in anesthetic concentration. At least 2 concentrations were
tested with similar results in 3 other preparations. The anesthetic
concentration for each wavelength is 0.5, 1, 1.5, and 4% from left to
right for each bar grouping. B: percent change in fluence rate at 0.5,
1, and 1.5% anesthesia relative to that measured at 4% anesthesia.
For the 514-nm wavelength, there is no change at any of the anesthetic concentrations tested relative to 4% anesthesia in the amount
of energy necessary to obtain a detectable response. For the 632- and
1,450-nm wavelengths, the fluence rate necessary to obtain a detectable response increases relative to the 4% anesthesia. For example,
at 1.5% anesthesia, 19% more energy is required to obtain a detectable response for 1,450 nm than at 0.5% anesthesia. Two trials were
done for the 632-nm wavelength and the data points are means ⫾
SD.
wavelength. However, the fractional change in fluence
rate necessary to reach a detectable response relative
to 4% anesthetic concentration varied with wavelength. Figure 2B illustrates that relative to 4% anesthesia, the 514-nm laser stimulus strength to reach
threshold was unchanged. However, for 632 and 1,450
nm, a stronger stimulus was required to reach a detectable response relative to 4% anesthesia. The effect
AJP-Regul Integr Comp Physiol • VOL
Fig. 3. Normalized response magnitude as a function of fluence rate
for different wavelengths at a stimulus duration of 1 s. Data for 400,
514, 632, and 1,450 nm are from 1 preparation. Data are means ⫾ SD
for 5 trials for each wavelength. These data were normalized to the
514-nm wavelength, which gave the largest response. Similar data
were obtained from 10 other preparations. Data for 10,600- and
3,390-nm wavelengths are from another preparation. Data are
means ⫾ SD for 3 trials for each wavelength. These data were
normalized to the 3,390-nm wavelength. Similar data were obtained
from 7 other preparations.
284 • FEBRUARY 2003 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
of anesthesia on threshold of the pit organ receptors
contrasts with that reported for halogenated anesthetics on A-delta and C-fiber nociceptive afferent fibers in
monkey where the threshold to a stimulus was decreased and response increased during halothane-N2O
anesthesia (9). Similar excitatory effects on A-delta
and C fiber nociceptors from the rabbit cornea were
seen with isoflurane, halothane, and enflurane (22). In
all subsequent experiments, the level of anesthesia
was kept constant at 1.5%.
Heart rate was slightly affected by the level of anesthesia, decreasing from 55 beats/min at 0.5% to 53
beats/min at 1 and 1.5% and 45 beats/min at 4%.
However, we do not believe that the decrease in response sensitivity as anesthetic dose increases can be
attributed to the change in heart rate. If as our data
suggest, the heat transfer agent is oxyhemoglobin and
water, a slower heart rate would increase the effectiveness of the heat transfer to the receptors and we would
expect a smaller stimulus strength necessary to each
threshold as anesthetic dose increases. We observed
just the opposite.
The response of the pit organ to a chemical stimulus
was tested. High-threshold heat receptors are particularly sensitive to capsaicin, the pungent element extracted from hot chili peppers (10). Neither a direct
application of capsaicin (50 ␮M in ethanol) to the pit
organ surface nor the vehicle ethanol had any effect on
the nerve response.
To determine the sensitivity of the pit organ afferents, we measured the nerve activity to different optical stimuli as a function of fluence rate (mW/cm2) and
wavelength. Figure 3 shows stimulus-response curves
for the receptive organ as reflected by the whole nerve
R602
SNAKE PIT ORGAN HEAT RECEPTORS
AJP-Regul Integr Comp Physiol • VOL
Fig. 4. Fluence rate necessary to produce a signal above background
for several wavelengths. The effectiveness of the different wavelengths to produce a detectable response follows precisely the stimulus-response curves shown in Fig. 3. Data are means ⫾ SD for 2
trials from 1 snake.
The response of many sensory receptors decreases
during a maintained stimulus due to a mechanism
called adaptation. To measure the magnitude of the
afferent activity as a function of stimulus duration, we
calculated the magnitude of the response for fixed time
intervals during and following cessation of the stimulus application. Because increasing stimulus strength
might recruit more receptors in our whole nerve recording and therefore mask any adaptive property of
the receptors, these experiments were made at fixed
stimulus strength. Figure 6, A-B, shows the response
magnitude during and following application of two
optical stimuli for two different durations, 1 and 10 s.
For the 1-s stimulus, the nerve firing record was divided into 200-ms bins and for the 10-s stimulus, it was
divided into 1-s bins. The response magnitude was
normalized to the maximum response for each stimulus wavelength. The nerve activity to the 1-s duration
stimulus reached maximum within 400 ms and was
maintained throughout and for the subsequent 200 ms
after the stimulus application ended. No adaptation
was apparent for this stimulus duration. However, the
response to the 10-s duration stimulus shown in Fig.
6B begins to decrease after 2 s even though the stimulus is maintained. The rate of the decrease is faster
for the least-effective 632-nm stimulus than it is for the
more effective 1,150-nm stimulus. As will be shown
below, the absorption of the longer wavelength by tissue water is better than that for the 632-nm stimulus
because of the higher coefficient of absorption of the
longer wavelength by water. It is expected, therefore,
that the 1,150-nm stimulus will be more effective compared with the 632-nm stimulus at activating the pit
organ receptors.
We measured the sensitivity of the pit organ from
the same preparation to four different wavelengths for
the same stimulus duration and power and compared
284 • FEBRUARY 2003 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
activity to optical stimuli of different wavelengths.
Wavelengths of 400; 514; 632; 1,450; 3,390; and 10,600
nm were used to irradiate the pit organ at stimulus
duration of 1 s and a fluence rate between 5 and 1,100
mW/cm2. From this graph, it can be seen that regardless of wavelength, increasing the strength of stimulus
resulted in an increase in response magnitude. However, the threshold for activation for the different response magnitudes varied depending on stimulus
wavelength. The responses to several wavelengths had
a tendency to show saturation with higher stimulus
powers. It is apparent that UV and IR, the extremes of
the electromagnetic wavelengths used in these studies,
are the most effective at activating the pit organ receptors. The wavelengths in the midrange of the spectrum
are least effective but nevertheless clearly activate the
pit organ receptors.
For both the 10,600- and 3,390-nm wavelengths,
increasing the stimulus magnitude past the point of
plateau sometimes resulted in an erratic or even decreasing response. These two wavelengths were very
effective at stimulating the pit organ receptors in the
copperhead and they produced almost identical stimulus-response curves. The 400-nm wavelength, the
shortest wavelength used, was the next most effective
wavelength. This wavelength was even more effective
than 1,450 nm, which is in the near IR. The leasteffective stimulus was 632 nm. There was often a
persistent discharge of the nerve following removal of
the stimulus at stimulus strengths near a saturating
response. This is consistent with heat transfer from
extracellular fluid to a temperature-sensitive plasma
membrane receptor, as it would be expected that heat
dissipation would be a function of stimulus duration
and strength.
We measured the minimum stimulus necessary to
elicit a nerve response from the pit organ receptors. As
Fig. 4 shows, the magnitude of the stimulus necessary
to reach threshold was found to be dependent on stimulus wavelength, with 3,390 nm being the most-effective stimulus and 632 nm being the least-effective
stimulus. The effectiveness of the stimulus correlates
well with the data shown in Fig. 3. Wavelengths, which
produce a steep increase in the nerve response, also
have the lowest threshold, whereas stimulus wavelengths that are less effective in activating the pit
organ receptors and that increase the nerve response
slowly or moderately have the highest threshold value.
The effect of stimulus duration for 514, 632, and
1,450 nm and three different stimulus strengths was
also investigated. Figure 5, A-C, shows a positive correlation between stimulus duration, stimulus strength,
and the magnitude of the nerve response for the three
different stimulus strengths. There is a consistent order in the effectiveness of the three wavelengths to
excite the receptor organ with 514 nm being the most
effective at all stimulus strengths and stimulus durations and 632 nm being the least effective. Similar
results were obtained in another copperhead in which
488 nm was substituted for 514 nm with the other two
wavelengths remaining the same.
R603
SNAKE PIT ORGAN HEAT RECEPTORS
Fig. 5. A-C: receptor response to 3 different wavelengths (514 nm, ■;
1,450 nm, ; 632 nm, F) as a function of increased stimulus duration
for 3 different fluence rates (A: 125, B: 280, and C: 1,025 mW/cm2).
All data are from the same preparation but the experiment was
repeated in a second snake.
the sensitivity with the absorptive properties of water
or oxygenated hemoglobin. Figure 7 shows the responses to four different optical stimuli. The sensitivity
of the pit organ has a positive correlation with the
absorption of these wavelengths by either water or
blood. As shown, the 400-nm stimulus produces the
largest response and 632 nm the smallest response.
The 400-nm wavelength is highly absorbed by oxyhemoglobin, and the 632-nm wavelength is only poorly
AJP-Regul Integr Comp Physiol • VOL
Fig. 6. A-B: pit organ receptors desensitize to a continuous stimulus.
A: nerve activity to 632 and 1,150 nm applied for 1 s and fluence rate
of 337.5 mW/cm2. The response is maintained for both wavelengths
throughout the stimulus duration. Maximal response for 632 nm is
258 ⫻ 10⫺6 V2/Hz ⫾ 12.5 ⫻ 10⫺6 SD, and for 1,150 nm it is 1,118 ⫻
10⫺6 V2/Hz ⫾ 33 ⫻ 10⫺6 SD. B: similar experiment except that the
stimulus duration is 10 s. The nerve activity begins to decrease after
2 s. Maximal response for 632 nm is 276 ⫻ 10⫺6 V2/Hz ⫾ 16 ⫻ 10⫺6
SD, and for 1,150 nm it is 1,211 ⫻ 10⫺6 V2/Hz ⫾ 177 ⫻ 10⫺6 SD. Each
point represents the mean ⫾ SD for 3–4 applications of the stimulus
from 1 preparation. This experiment was repeated in 4 snakes with
similar results. F, 632 nm; ■, 1,150 nm.
284 • FEBRUARY 2003 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
absorbed by this chromophore. Figure 8 shows the
absorption spectra for oxyhemoglobin and water with
the seven optical stimuli used in these studies indicated on the graph (15, 18). Above 1,200 nm, the
absorption spectra of oxygenated and deoxygenated
blood are not significantly different and follow the
water absorption curve (31). The sensitivity of the pit
organ receptors to electromagnetic radiation follows
closely the absorption spectra of vascularized tissue.
The 632-nm wavelength is poorly absorbed by oxyhemoglobin or water and is the least-effective stimulus.
The 400-, 3,390-, and 10,600-nm wavelengths are the
most-effective stimuli and are well absorbed by oxyhemoglobin or water. The 488-, 514-, and 1,450-nm wavelengths are intermediate in their ability to excite the
receptor organ and generate nerve activity.
R604
SNAKE PIT ORGAN HEAT RECEPTORS
DISCUSSION
Our finding that halogenated anesthesia reduces the
ability of the pit organ receptors to generate nerve
activity is unlike that shown for similar anesthetics on
mammalian small myelinated and unmyelinated fibers. Although this finding is not germane to the main
focus of this study, it demonstrates the importance of
maintaining a constant anesthetic level so that the
response characteristics of the pit organ receptor response to different optical stimuli can be compared.
The nerve innervating the pit organ consists of small
myelinated fibers similar to the A␦ group in mammals.
In mammals, this fiber type is associated with nociceptors including thermal and mechanically sensitive receptors that produce sensations of pain (30). Studies on
Fig. 8. Absorption spectra for oxyhemoglobin and water. The absorption
spectrum for oxyhemoglobin is from
Gordy and Drabkin (15), and the absorption spectrum for water is from
Hale and Querry (18). The vertical
lines indicate the optical stimuli used
in these experiments. Right: depth of
penetration in water or oxygenated
blood for the bandwidth shown. These
are simply the reciprocal of the absorption coefficients.
AJP-Regul Integr Comp Physiol • VOL
284 • FEBRUARY 2003 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
Fig. 7. Response of the pit organ afferents to 4 different wavelengths
(400; 488; 1,450; and 632 nm). Stimulus duration was 1 s; fluence
rate for all stimuli was 90 mW/cm2. The number above each vertical
bar represents the absorption coefficient of that wavelength in water
or oxygenated blood as indicated by the subscripted identifier (w) or
(b).
the sensitivity of cutaneous nociceptors responding to
both heat and mechanical stimulation in mammals to
halothane anesthesia have demonstrated an increase
in sensitivity to heat but not mechanical stimulation
(9). Similar results were reported for mechanoreceptors innervating the rabbit cornea (22). The mechanism for the increased sensitivity is unknown but different mechanisms may involve ion channels targeted
by a specific anesthetic (21, 22). The nerve fibers innervating the temperature receptors of the pit organ
belong to the same class as the mammalian nociceptors. However, they do not respond to mechanical stimulation and are sensitive to temperatures in the nonnoxious range. Their sensitivity to halogenated
anesthetics may be governed by yet other membrane
mechanisms. The pit organ receptive area is supplied
by a dense capillary network that may function as a
heat exchange mechanism (1, 2, 3). The sensitivity of
the receptors to anesthetic may reflect this dense vascularization of the pit organ tissue.
Our results do not support the hypothesis that a
photosensitive protein is responsible for sensing the
electromagnetic radiation. Photoreceptors express photosensitive pigments with a half-maximal sensitivity of
⬃100 nm and maximum sensitivity with a single peak.
This compares to the multiple-peak, wide-band spectral tuning of the pit organ receptors. If the pit organ
receptors were similar to the photosensitive proteins
that transduce electromagnetic radiation into an electrical signal in photoreceptors, the nerve terminals
would have to express several photosensitive pigments. Evidence suggests that no opsin-like proteins
similar to those expressed in vertebrate photoreceptors
are expressed in the pit organ (16). The nerve terminals supplying the pit organ contain many mitochondria. The peak absorption of cytochrome c is near 400
nm, and we cannot discount the possibility that this
protein in mitochondria is the heat transfer agent at
this particular wavelength. However, this single pro-
R605
SNAKE PIT ORGAN HEAT RECEPTORS
AJP-Regul Integr Comp Physiol • VOL
Because of the complex signaling cascade associated
with the G protein-coupled photosensitive pigment in
photoreceptors, a response that outlasts the stimulus
is observed when rhodopsin molecules absorb only a
few photons (4). However, if tissue heating were responsible for receptor activation, it would be expected
that the removal of heat from the pit membrane would
depend, in part, on the radiative properties of the
tissue and the removal of heat by blood flowing through
the pit membrane. Thus, the stronger stimulus results
in a greater increase in tissue temperature and a
longer lasting response as observed in our experiments. The pit membrane is highly vascularized (2)
and it has been suggested that this functions as a heat
removal system for the pit organ. It has been shown
that local heating or the application of capsaicin can
alter blood flow in human skin at the stimulated site
(33).
Heat detection in mammals is mediated by receptors
expressed in peripheral nerve terminals. A highthreshold heat receptor sensitive to capsaicin, the pungent extract from hot chili peppers, called vanilloid
receptor type I (VR1), has been cloned from a rat dorsal
root ganglion library (10). This receptor has a high
threshold to thermal stimulus (⬎40°C) and besides its
sensitivity to capsaicin, it can also be activated by
hydrogen ions (10). This receptor is thought to mediate
temperature-induced pain. Low-threshold temperature receptors have yet to be cloned from the mammalian nervous system. It is possible that low-threshold
temperature receptors, so called warm receptors, may
belong to the same family as VR1. However, we were
unable to activate the snake pit receptors by the application of capsaicin directly to the pit membrane at
concentrations that would produce strong stimulation
of the mammalian VR1. In addition, Bullock and
Diecke (7) applied mildly acidic solutions to the pit
organ and failed to elicit nerve activity. However,
whether or not the pH actually was altered at the
receptor is unknown. Low pH has been shown to shift
the temperature sensitivity of VR1 to lower temperatures (10). The absence of a response to VR1 agonists
suggests that the protein expressed in the pit membrane is not a nociceptor and is consistent with its
sensitivity to temperatures that would produce sensations of warmth.
VR1 is expressed in the cell bodies of dorsal root
ganglia. As has been demonstrated for VR1, it might be
expected that the receptors expressed in the nerve
terminals of the pit organ would also be expressed in
the trigeminal ganglion cells from which these axons
arise. We recorded from primary cultures of the snake
trigeminal ganglion cells and observed membrane currents associated with temperature changes in the
range of 15–30°C (unpublished results). In addition,
we demonstrated a marked shift toward lower temperatures of a modified VR1 (10a) demonstrating that an
integral membrane protein can act as a low-temperature receptor. This modified VR1, made by removing
the NH2 terminus, has a temperature threshold of
⬃30°C.
284 • FEBRUARY 2003 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
tein cannot account for the wide-band spectral tuning
of the pit organ receptors.
The major finding is that the pit organ receptors are
sensitive to a wide range of electromagnetic radiation
with peak sensitivities found for specific wavelengths
over the entire range. Furthermore, the strongest response was observed for radiation at wavelengths that
are strongly absorbed by either oxyhemoglobin or water content of the pit membrane. In the near UV and
visible range (400–1,000 nm), the major absorbers in
biological tissue are hemoglobin and melanin of the
skin (19a). Above 1,200 nm, photons are absorbed primarily by water or blood, which have similar absorbance properties (19a, 31). It has been reported that
the pit membrane is a highly perfused tissue with
unusual capillary density (2). Thus, a small amount of
optical radiation can alter local tissue temperature. In
the IR region, it is well known that water molecules in
tissue strongly absorb optical radiation resulting in
local temperature increases in tissue (18). The spectral
sensitivity of the pit organ receptors covers a range
from at least 400 to 10,600 nm, which is far greater
than would be expected for opsin-based photopigments.
Therefore, it is unlikely that receptor activation in the
pit organ requires a photochemical process similar to
that found in photoreceptors. Instead, the pit organ
sensory tissue is likely to be a set of thermal receptors
that exhibit the highest sensitivity to wavelengths
where the absorption and consequently local heat generation are the strongest. Their tuning is, therefore,
dependent on the absorptive properties of the surrounding tissue. The infrared receptive organ in pit
vipers is only 15–25 ␮m thick measured in fixed tissue
and consists of neural and capillary tissues sandwiched between two layers of epithelium (5, 8, 28, 29).
The pit membrane is located below the surface of the
surrounding tissue at ⬃1-mm depth. This further isolates the structure from air currents. A similar structure has been described in some beetles with thermally
sensitive receptive organs (26). The small thermal
mass of the pit organ surrounded by insulating layers
of air provides an ideal site for rapid and localized
heating of the tissue and thus stimulation of receptors
expressed in free nerve endings. This is in contrast to
mammalian thermal receptors that are distributed
within the deeper layers of tissue and surrounded by a
larger thermal mass.
The nerve response to direct illumination of the pit
organ with the different optical stimuli increases
monotonically with stimulus strength and stimulus
duration until maximal nerve firing rate is reached.
Both the threshold for activation and the sensitivity of
the pit organ receptors to the different optical stimuli
are in the order of 10,600 ⫽ 3,390 ⬎ 400 ⬎ 514 ⬎
1,450 ⬎ 632 nm. This is the same order as the combined absorption spectra of oxyhemoglobin and water.
Our observation that only the response to intense
optical radiation can produce an afterdischarge when
the stimulus is removed supports our hypothesis that
the pit receptors are sensitive thermal receptors. Photoreceptors do not have this response characteristic.
R606
SNAKE PIT ORGAN HEAT RECEPTORS
In conclusion, our evidence strongly supports the
hypotheses that receptors in the pit organ of pit vipers
are temperature receptors that are spectrally tuned
based on the absorptive properties of the surrounding
tissue and suggests that unique receptors with high
sensitivity to ambient temperature are expressed in
nerve terminals ending in the pit membrane enabling
this organ to exhibit a highly sensitive and unique
response to a thermal stimulus.
This work was supported in part by Multidisciplinary University
Research Initiative grant from the Air Force Office of Sponsored
Research, the Department of Defense.
1. Amemiya F, Goris RC, Masuda Y, Kishida R, Atobe Y, Ishii
N, and Kusunoki T. The surface architecture of snake infrared
receptor organs. Biomed Res (Tokyo) 16: 411–421, 1995.
2. Amemiya F, Nakano M, Goris RC, Kadota T, Atobe Y,
Funakoshi K, Hibiya K, and Kishida R. Microvasculature of
crotaline snake pit organs: possible function as a heat exchange
mechanism. Anat Rec 254: 107–115, 1999.
3. Amemiya F, Ushiki T, Goris RC, Atobe Y, and Kusunoki T.
Ultrastructure of the crotaline snake infrared pit receptors: SEM
confirmation of TEM findings. Anat Rec 246: 135–146, 1996.
4. Baylor DA. Photoreceptor signals and vision. Invest Ophthalmol Vis Sci 28: 34–49, 1987.
5. Bleichmar H and de Robertis E. Submicroscopic morphology
of the infrared receptor of pit vipers. Z Zellforsch Mikrosk Anat
56: 748–761, 1962.
6. Brown PK and Wald G. Visual pigments in human and monkey retinas. Nature 200: 37–43, 1963.
7. Bullock TH and Diecke FPJ. Properties of an infrared receptor. J Physiol 134: 47–87, 1956.
8. Bullock TH and Fox W. The anatomy of the infrared sense
organ in the facial pit of pit vipers. Q J Microsc Sci 98: 219–234,
1957.
9. Campbell JN, Raja SN, and Meyer RA. Halothane sensitizes
cutaneous nociceptors in monkeys. J Neurophysiol 52: 762–770,
1984.
10. Caterina MJ, Schumacher MA, Tominaga M, Rosen TA,
Levine JD, and Julius D. The capsaicin receptor: a heatactivated ion channel in the pain pathway. Nature 389: 816–824,
1997.
10a.Christensen BN, Moiseenkova V, Hellmich H, and Motamedi M. Functional characterization of a truncated form of
VR1/eGFP fusion protein (Abstract). Society for Neuroscience
2001.
11. De Cock Buning T. Thermal sensitivity as a specialization for
prey capture and feeding in snakes. Am Zool 23: 363–375, 1983.
12. De Cock Buning T. Thresholds of infrared sensitive tectal
neurons in Python reticulatus, Boa constrictor and Agkistrodon
rhodostoma. J Comp Physiol [A] 151: 461–467, 1983.
13. De Cock Buning T, Terashima S, and Goris RC. Crotaline
pit organs analyzed as warm receptors. Cell Mol Neurobiol 1:
69–95, 1981.
AJP-Regul Integr Comp Physiol • VOL
284 • FEBRUARY 2003 •
www.ajpregu.org
Downloaded from http://ajpregu.physiology.org/ at Univ of California-San Diego on May 25, 2013
REFERENCES
14. De Cock Buning T, Terashima S, and Goris RC. The role of
thermosensitivity in the feeding behavior of the pit viper Agkistrodon blomhoffi brevicaudus. Jpn J Herpetol 9: 7–27, 1981.
15. Gordy E and Drabkin DL. Spectrophotometric studies. XVI.
Determination of the oxygen saturation of blood by a simplified
technique, applicable to standard equipment. J Biol Chem 227:
285–299, 1957.
16. Grace MS, Church DR, Kelly CT, Lynn WF, and Cooper
TM. The Python pit organ: imaging and immunocytochemical
analysis of an extremely sensitive natural infrared detector.
Biosens Bioelectron 14: 53–59, 1999.
17. Grace MS, Woodward OM, Church DR, and Calisch G. Prey
targeting by the infrared-imaging snake Python molorus: effects
of experimental and congenital visual deprivation. Behav Brain
Res 119: 23–31, 2001.
18. Hale GM and Querry MR. Optical constants of water in the
200 nm to 200 ␮m wavelength region. Applied Optics 12: 555–
563, 1973.
19. Harris JF and Gamow RI. Snake infrared receptors: thermal
or photochemical mechanism? Science 172: 1252–1253, 1971.
19a.Jacques SL. Modelling light propagation in tissues. In: Biomedical Optical Instrumentation and Laser-Assisted Biotechnology,
edited by Verga Scheggi AM, Martellucci S, Chester AN, Pratesi
R. Boston, MA: Kluwer Academic, 1996.
20. Klauber LM. Rattlesnakes: Their Habits, Life Histories and
Influence on Mankind (2nd ed.). Berkeley: Univ. of California
Press, 1972.
21. MacIver MG and Kendig JJ. Anesthetic effects on resting
membrane potential are voltage-dependent and agent-specific.
Anesthesiology 74: 83–88, 1991.
22. MacIver MB and Tanelian DL. Volatile anesthetics excite
mammalian nociceptors afferents recorded in vitro. Anesthesiology 72: 1022–1030, 1990.
24. Newman EA and Hartline PH. The infrared “vision” of
snakes. Sci Am 246: 116–128, 1982.
25. Noble GK and Schmidt A. The structure and function of the
facial and labial pits of snakes. Proc Am Philos Soc 77: 263–288,
1937.
26. Schmitz H and Bleckmann H. Fine structure and physiology
of the infrared receptor of beetles of the genus Melanophila
(Coleoptera:Buprestidae). Int J Insect Morph Embryol 26: 205–
215, 1997.
28. Terashima S, Goris RC, and Katsuki Y. Structure of warm
fiber terminals in the pit membrane of vipers. J Ultrastruct Res
31: 494–506, 1970.
29. Terashima S, Goris RC, and Katsuki Y. Generator potential
of crotaline snake infrared receptor. J Neurophysiol 31: 682–
688, 1968.
30. Willis WD. The pain system. The neural basis of nociceptive
transmission in the mammalian nervous system. Pain Headache
8: 1–346, 1985.
31. Roggan A, Friebel M, Dörschel K, Hahn A, and Müller G.
Optical properties of circulating human blood in the wavelength
range 400–2500 nm. J Biomed Opt 4: 36–46, 1999.
33. Stephens DP, Charkoudian N, Benevento JM, Johnson
JM, and Saumet JL. The influence of topical capsaicin on the
local thermal control of skin blood flow in humans. Am J Physiol
Regul Integr Comp Physiol 281: R894–R901, 2001.