behavioral model of acute pain paws in thermoneutral ... - nocions

Thermoregulatory vasomotor tone of the rat tail and paws in thermoneutral conditions and its impact on a behavioral model of acute pain

Nabil El Bitar, Bernard Pollin, Elias Karroum, Ivanne Pincedé, André Mouraux and Daniel Le Bars J Neurophysiol 112:2185-2198, 2014. First published 9 July 2014; doi:10.1152/jn.00721.2013 You might find this additional info useful... Supplemental material for this article can be found at: /content/suppl/2014/07/17/jn.00721.2013.DC1.html This article cites 83 articles, 33 of which can be accessed free at: /content/112/9/2185.full.html#ref-list-1 This article has been cited by 1 other HighWire hosted articles ''On-'' and ''off-'' cells in the rostral ventromedial medulla of rats held in thermoneutral conditions: are they involved in thermoregulation? Nabil El Bitar, Bernard Pollin and Daniel Le Bars J Neurophysiol, November 1, 2014; 112 (9): 2199-2217. [Abstract] [Full Text] [PDF]

Additional material and information about Journal of Neurophysiology can be found at: http://www.the-aps.org/publications/jn

This information is current as of November 4, 2014.

Journal of Neurophysiology publishes original articles on the function of the nervous system. It is published 12 times a year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2014 by the American Physiological Society. ISSN: 0022-3077, ESSN: 1522-1598. Visit our website at http://www.the-aps.org/.

Downloaded from on November 4, 2014

Updated information and services including high resolution figures, can be found at: /content/112/9/2185.full.html

J Neurophysiol 112: 2185–2198, 2014. First published July 9, 2014; doi:10.1152/jn.00721.2013.

Thermoregulatory vasomotor tone of the rat tail and paws in thermoneutral conditions and its impact on a behavioral model of acute pain Nabil El Bitar,1,2 Bernard Pollin,1,2 Elias Karroum,1,2 Ivanne Pincedé,1,2 André Mouraux,3 and Daniel Le Bars1,2 1

Sorbonne Universités, Université Pierre et Marie Curie, Faculté de Médecine Paris, France; 2Neurosciences Paris-Seine, Institut National de la Santé et de la Recherche Médicale UMRS-1130, Centre National de la Recherche Scientifique UMR-8246, Paris, France; and 3Institute of Neuroscience, Université Catholique de Louvain, Brussels, Belgium

Submitted 7 October 2013; accepted in final form 8 July 2014

thermoneutrality; blood pressure variability; heart-rate variability; vasomotion; pain tests and models

on the physiology, pathophysiology, and pharmacology of pain relies on an assessment of the “behavioral” responses of rodents following the delivery of a thermal nociceptive stimulus to the tail or paw [reviewed in Le Bars et al. (2001, 2009)]. For example, in the “tail-flick test” (d’Amour and Smith 1941), i.e., one of the most widely used procedures to assess pain in rodents, a thermal stimulus is delivered to the tail, and the latency of tail withdrawal is used as a measure inversely related to “pain sensitivity.” Consequently, increases and decreases in the latency of tail withdrawal are interpreted as hypo- and hyperalgesia, respectively. However, the tail of the rat plays a crucial role in thermoregulation. The role of the tail as a radiator (e.g., heat dissipation by increasing blood flow and the inverse) was first described by Knoppers (1942) and has been largely confirmed

A LARGE AMOUNT OF RESEARCH

Address for reprint requests and other correspondence: D. Le Bars, Faculté de Médecine, UPMC, 91 Boulevard de l’Hôpital, 75013 Paris, France (e-mail: [email protected]). www.jn.org

(Dawson and Keber 1979; Grant 1963; Hellström 1975; Little and Stoner 1968; O’Leary et al. 1985; Raman et al. 1983; Rand et al. 1965; Vanhoutte et al. 2002; Young and Dawson 1982). The tail lacks fur, has a large surface area-to-volume ratio, is well vascularized (Gordon 1990, 1993), and accounts for ⬃7% of the total surface area of the body (Lin et al. 1979). The dissipation of heat is regulated by abrupt “on-off” variations of blood flow in a system of arterial-venous anastomoses that form a double ladder along the tail (Dawson and Keber 1979; Gemmell and Hales 1977; Rand et al. 1965; Thorington Jr. 1966; Young and Dawson 1982). When open, blood flow can increase by a factor of 35– 40 (Aukland and Wiig 1984; Nakajima et al. 1999; Raman et al. 1983; Vanhoutte et al. 2002) and lead to increases in tail skin temperature (Tskin) of ⬎10°C (O’Leary et al. 1985). The involvement of the paws in rat thermoregulation was first mentioned by Grant (1963). The feet make up ⬃10% of the total surface area of the body, slightly more than the tail (Lin et al. 1979). Skin on the nonhair-covered plantar surface of the paws exhibits both a high proportion of arterioles and venules and a high basal and heat-stimulated blood flow (Rendell et al. 1993, 1998). The feet vasodilate in near synchrony with the tail during ambient heat stress, internal heating caused by exercise, or drug-induced hyperthermia (Gordon and Watkinson 1988; Key and Wigfield 1992, 1994; Lin et al. 1978, 1979; Thompson and Stevenson 1965). Although most often not considered, or ignored, in pain research, these thermoregulatory mechanisms are likely to have a critical effect on the responses triggered by thermal nociceptive tests. We recently developed and validated in the rat, the mouse, and the human a psychophysical approach to characterize nociceptive responses triggered by heating the skin (Benoist et al. 2008; Pincedé et al. 2012; Plaghki et al. 2010). This approach is based on the joint analysis of the stimulus and the response, including the measurement of three observable variables, namely the initial Tskin, the heating rate, and the temperature reached at the actual moment of the response. We showed that the behavioral latency to radiant heat applied to the tail of rats or mice is inversely related to the Tskin (Benoist et al. 2008; Pincedé et al. 2012). This relation results from considerable variations of the conduction velocity of the peripheral fibers that elicit the behavioral response, itself dependent on the temperature of the tail. When the temperature of the tail rises from 20°C to 34°C, the conduction velocity of the fibers can be increased by a factor of three. Furthermore, the threshold to elicit a withdrawal response increases by 4°C,

0022-3077/14 Copyright © 2014 the American Physiological Society

2185


El Bitar N, Pollin B, Karroum E, Pincedé I, Mouraux A, Le Bars D. Thermoregulatory vasomotor tone of the rat tail and paws in thermoneutral conditions and its impact on a behavioral model of acute pain. J Neurophysiol 112: 2185–2198, 2014. First published July 9, 2014; doi:10.1152/jn.00721.2013.—The tail and paws in rodents are heat exchangers involved in the maintenance of core body temperature (Tcore). They are also the most widely used target organs to study acute or chronic “models” of pain. We describe the fluctuations of vasomotor tone in the tail and paws in conditions of thermal neutrality and the constraints of these physiological processes on the responses to thermal nociceptive stimuli, commonly used as an index of pain. Skin temperatures were recorded with a calibrated thermal camera to monitor changes of vasomotor tone in the tail and paws of awake and anesthetized rats. In thermoneutral conditions, the sympathetic tone fluctuated at a rate of two to seven cycles/h. Increased mean arterial blood pressure (MAP; ⬃46 mmHg) was followed by increased heart rate (HR; ⬃45 beats/min) within 30 s, vasoconstriction of extremities (3.5–7°C range) within 3–5 min, and increased Tcore (⬃0.7°C) within 6 min. Decreased MAP was followed by opposite events. There was a high correlation between HR and Tcore recorded 5– 6 min later. The reaction time of the animal’s response to a radiant thermal stimulus— heat ramp (6°C/s, 20 mm2 spot) generated by a CO2 laser— directed to the tail depends on these variations. Consequently, the fluctuations in tail and paw temperature thus represent a serious confound for thermal nociceptive tests, particularly when they are conducted at thermal neutrality.

2186

INTERACTIONS BETWEEN THERMOREGULATION AND NOCICEPTION

possibly because the response is dependent on the contrast between baseline and target Tskin. Therefore, fluctuations in Tskin could be a major source of variation, potentially biasing the results of experiments studying pain and nociception using thermal stimuli delivered to the rat tail or paw [see also Han and Ren (1991); Hole et al. (1990); Le Bars et al. (2001, 2009); Tjølsen and Hole (1997)]. The aims of the present study were to characterize in thermoneutral condition (Gordon 1993; Romanovsky et al. 2002), i.e., the condition that is usually considered as the best on homeostatic ground and therefore, most suited to study nociception, the following: 1) the dynamics of the thermoregulatory mechanisms controlling the vasomotor tone and Tskin of the rat tail and paw; 2) the correlation among tail and paw Tskin, core body temperature (Tcore), mean arterial blood pressure (MAP), and heart rate (HR); and 3) the impact of these thermoregulatory mechanisms on the behavioral response triggered by a thermonociceptive stimulus applied to the tail. Importantly, as general anesthetics are known to alter thermoregulation, the experiments were performed in both awake and anesthetized rats to establish standards for optimization of the experimental setup in anesthetized animals. ⌬HLI ⌬HLItail-dist ⌬HLItail-mid ⌬HLItail-prox ⌬HR ⌬MAP ⌬T ⌬Tcore ⌬TROI ⌬Tskin BAT BRAC FFT HLI HLImax HLImin HLIpaw-left HLIpaw-right HLIROI HLItail-dist HLItail-mid HLItail-prox HR LF MAP paw-left paw-right

Variation of heat-loss index ⌬HLI of the distal part of the tail ⌬HLI of the middle part of the tail ⌬HLI of the proximal part of the tail Variation of heart rate (beats/min) Variation of mean arterial blood pressure [millimeter of mercury (mmHg)] Variation of temperature (°C) Variation of core body temperature (°C) Variation of region of interest (where the temperature was monitored through thermoimaging) temperature Variation of skin temperature (°C) Brown adipose tissue Ultradian basic rest-activity cycles Fast Fourier transform Heat-loss index Maximal value of HLI Minimal value of HLI HLI of the plantar aspect of the left hind paw HLI of the plantar aspect of the right hind paw HLI of region of interest (where the temperature was monitored through thermo-imaging) HLI of the distal part of the tail HLI of the middle part of the tail HLI of the proximal part of the tail Mean heart rate (beats/min) Low frequency (as defined in the mean arterial blood pressure/HR variability literature) Mean arterial blood pressure (mmHg) Mid-plantar area on the left hind paw Mid-plantar area on the right hind paw

ROI Tambient Tcore Tpaw-left Tpaw-right TROI Tskin Ttail-dist Ttail-mid Ttail-prox Twarm tail-dist tail-mid tail-prox TFL VLF

Maximum Bravais-Pearson correlation coefficient Region of interest (where the temperature was monitored through thermo-imaging) Ambient temperature (°C) Core body temperature (°C) Temperature of the plantar aspect of the left hind paw (°C) Temperature of the plantar aspect of the right hind paw (°C) ROI temperature Temperature of the skin (°C) Temperature of the distal part of the tail (°C) Temperature of the middle part of the tail (°C) Temperature of the proximal part of the tail (°C) Warming temperature (°C) Distal area of the tail, located 3 cm from the tip Intermediate area of the tail, located at mid-tail Proximal area of the tail, located 3 cm from the root Tail-flick latency Very low frequency (as defined in the MAP/HR variability literature)

METHODS

Ethical Statement Animal experiments were performed with permission of the Board of the Veterinarian Services of the French Ministry of Agriculture (permit number 75-151) in accordance with the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals, the European Communities Council Directive 86/609/EEC regulating animal research, and the Ethics Committee of the International Association for the Study of Pain (Covino et al. 1980; Zimmermann 1983). Procedures were approved by the Ethics Committee for Animal Experimentation of our institution. Animals A total of 87 adult male Sprague-Dawley rats (Janvier Labs, Saint-Berthevin, France) took part in the experiments, weighting 320 –370 g. They were housed in groups of three to four rats/cage, allowed free access to food and water and a 12-h alternating light-dark cycle, and acclimatized to the laboratory for at least 1 wk before the experiments. The experiments were conducted between 9 AM and 5 PM. Experimental Procedures in Awake Rats Seven awake animals were placed in a conventional restrainer (LE5024; Bioseb, Chaville, France). The tail was inserted in five anchored, aligned steel rings, maintained in place during the entire experiment. When the animal was stressed, for any reasons, the tail was vasoconstricted, as expected (Carrive et al. 2011; Vianna et al 2008). The animals were habituated to the testing conditions (one to two trials). In fact, the avoidance of stress was easy to achieve on the essential condition that the Tambient allowed the thermoneutrality of the animal (checked by the thermographic recording; see RESULTS). At

J Neurophysiol • doi:10.1152/jn.00721.2013 • www.jn.org


Glossary

rBP


Experimental Procedures in Anesthetized Rats

In 15 experiments, the variations of skin blood flow were measured at the right heel by means of a conventional laser Doppler probe (DRT4; Moor Instruments, Axminster, UK) and compared with the ⌬Tskin measured at the same site. There was a strong correlation between the laser Doppler measure of skin blood flow and the thermographic measure of surface Tskin (Fig. 1). The temperature recordings provided a smoothed, slightly delayed, but nevertheless faithful, image of the underlying skin blood flow (Hertzman 1953). This was confirmed by the cross-correlation analysis of the data measured with a 1-second temporal resolution [mean lag of 35 s (13–57 s); rBP ⫽ 0.73 (0.68 – 0.78)]. Core and Heating Temperature A two-channel Omega HH506RA digital thermometer and two VIP-T-CT25515 probes (0.1°C resolution) were used to measure the

A

Thermographic Measures of Tskin

3

0

-3

MAP and HR The MAP and HR were monitored continuously via the catheter inserted into the common carotid artery, using a transducer connected to a computer. The MAP and HR were calculated and recorded using the NOTOCORD blood pressure analyzer system. Measures were averaged every 60 s.

0

B temperature of the paw (z-score)

An infrared camera (Jade MWIR; Cedip Infrared Systems, CroissyBeaubourg, France), with a 3- to 5-␮m optical bandpass and a 500-␮s integration time, was used to obtain images of 320 ⫻ 240 pixels at a 1-Hz sampling rate, with a sensitivity of 0.02°C at 25°C. The camera was placed 1.5 m above the scene and was controlled by Cirrus software (Cedip Infrared Systems). The camera was calibrated using a black body, as described previously (Benoist et al. 2008). Altair software (Cedip Infrared Systems) was used to explore the spatial and temporal dynamics of the Tskin at the level of the tail and paws as follows. First, several ROIs were defined in the recorded scene, each comprising 10 pixels. Three ROIs were located on the tail. A proximal ROI (tail-prox) was placed 3 cm from the root of the tail, an intermediate ROI (tail-mid) was placed at the middle of the tail, and a distal ROI (tail-dist) was placed 3 cm from the tip of the tail. Two additional ROIs were located on the plantar aspect of the left and right hind paws of the anesthetized rats (paw-left and paw-right). Finally, an ROI was located over a piece of wood placed in the scene to monitor the Tambient. For each time point, the mean of the 10 pixels defining each ROI was computed to obtain one single temperature time course for each ROI (Ttail-prox, Ttailmid, T tail-dist, T paw-left, T paw-right, and T ambient).

10

20

min

5

3

1

-1

-3

y = 0,83*x; F(1-3900) = 8701; R2 = 0,69; p< 0,0001 -5 -5

-3

-1

1

3

5

laser-Doppler (z-score) Fig. 1. ⌬Tskin reflects the variations of the underlying blood flow. The data are expressed in z-score. A: example of simultaneous recording of the blood flow measured through a conventional laser Doppler probe (black curve) and the temperature (yellow curve) at the same site during a 20-min sequence (1second temporal resolution). The position of the laser Doppler probe is shown on the picture in the upper-right corner of Fig. 4. ⌬Tskin appeared as a smoothed image of the laser Doppler signal. B: relationship between the blood flow measured by laser Doppler and the Tskin. Note the positive linear relation (15 experiments).



Seventy-two animals were deeply anesthetized with 2.5% halothane in 100% oxygen. A tracheal cannula was inserted, and the ventilation was controlled mechanically with an open-circuit respirator, equipped with a scavenging system at a rate of 50 counts/min. The expiratory halothane level and end-tidal CO2 were assessed with a capnometer (Capnomac II; Datex Instruments, Helsinki, Finland), sampled every 10 s and under control of alarms throughout the experiment. A catheter was inserted into the common carotid artery to monitor MAP and HR After surgery, halothane was optimized to maintain anesthesia without troubling thermoregulation and tail vasomotor fluctuations. The mean level was 0.81% (0.79 – 0.82%). Oxygen was kept at 100%. Tidal volume was adjusted to keep the end tidal CO2 at 3.6% (ranging from 3.4% to 3.7%). Experimental measures started at least 30 min after surgery. At the end of the experiment, the rat was killed with an intraperitoneal thiopental injection. In a few experiments, we were able to replicate the experiments described in the present paper with isoflurane (0.9 –1.1%); quantitatively, the variations of the variables were similar and in the same range. To maintain the anesthetized rat in a thermoneutral condition, the body was wrapped up with a water-warming pad connected to an extracapacity warm-water circulator (TP-220; Kent Scientific, Torrington, CT), sparing the head, paws, and tail. The heating blanket was covered with an isothermic, metalized, polyester film to stabilize the temperature around the body. Twarm was adjusted to 0 – 0.3°C above Tcore after stabilization [38.2°C (38.0 –38.3°C)]. An air conditioner was used to maintain a stable ambient room temperature [24.1°C (23.9 –24.3°C)].

Laser Doppler Measure of Skin Blood Flow

z-score

the end of the experiment, the rats were killed with an intraperitoneal thiopental injection.

2187

2188


Tcore. The probe was inserted 6 cm in the rectum (Lomax 1966). The second probe was placed around the trunk of the rat, such as to estimate the heating provided by the isothermal warming blanket. Each measurement was sampled every 60 s. Heat Loss Index Acting as an interface between the body core and the external environment, the skin of the tail and paws operates as radiators. Therefore, the temperature of the skin is dependent not only on the vasomotor tone of the skin vessels but also on the Tambient and the Tcore. To allow a direct comparison of the results obtained in the different rats, for each ROI and time-point, we calculated the HLI, as proposed by Székely (1986): HLIROI ⫽ (TROI ⫺ Tambient)/(Tcore ⫺ Tambient). The obtained index varies between zero (when the Tskin equals the Tambient) and one (when the Tskin equals the Tcore) (Gordon et al. 2002; Romanovsky and Blatteis 1996; Romanovsky et al. 2002). Experimental Procedures in Awake Rats for Tail-Flick Test

A

Data Processing and Statistical Analysis Only recordings lasting ⬎60 min were considered. The recordings were analyzed using Matlab R2006a (MathWorks, Natick, MA) to compute the following: 1) Descriptive statistics. Data were represented as mean with their 95% confidence interval.

P

55

B

60

65

70

time(min)

38

temperature(°C)

Ttail-prox Ttail-mid

34

Ttail-dist Tambient

30

26

C Tcore (°C)

38

37 0

20

40

60

80

100

120

140

160

180

time(min)

Fig. 2. Individual example of a 3-h thermographic recording of the spontaneous variations of tail temperature in an awake rat. A: colored pictures are taken every minute from a 20-min sequence of a thermographic recording (indicated by the vertical gray background in B and C). The 5 steel rings securing the tail are seen as dark horizontal lines. B: Tskin time courses were obtained by averaging the temperatures within 3 ROIs, located at the Ttail-prox, Ttail-mid, and Ttail-dist. An additional ROI was located over a piece of wood, measuring the Tambient. Note the important fluctuations in Tskin, varying between temperatures close to Tambient (vasoconstricted state) and Tcore (vasodilated state) during the augmentation of Tambient. C: the graph represents Tcore, measured using a rectal probe. J Neurophysiol • doi:10.1152/jn.00721.2013 • www.jn.org


Eight awake rats were habituated, as described above, and maintained in thermoneutral conditions, except that the day before the experiment, the tail was depilated with a depilatory cream (DermoTolerance hair-removal cream; Vichy Laboratories, Cusset, France), as described previously (Benoist et al. 2008; Pincedé et al. 2012). The day of the experiment, each animal was placed in a conventional restrainer (LE5024; Bioseb). At the end of the experiment, the rats were killed with an intraperitoneal thiopental injection.

The tail-flick test was performed as follows. A temperature-controlled CO2 laser stimulation device (Lug, Ferrières, Belgium) was used to heat a small spot of the skin of the tail using a ramp, increasing linearly from the initial temperature at a constant 6°C/s rate. Beam diameter at target site was 6 mm. Importantly, the laser beam of this stimulator has a flat-top distribution profile, such that the Tskin is homogeneous within the entire stimulated area. The laser stimuli were delivered to the right or left sides of the middle of the tail (beam oriented at 45° with respect to the vertical) to elicit a contralateral withdrawal movement of the tail. The target of the stimulus was shifted slightly after each trial. The intertrial interval at a same site was at least 15 min. The radiant heat was applied to the middle of the tail, and the Tskin was measured on two points—1.5 cm on both sides of the laser target—themselves, slightly shifted from a given stimulus to the next. The mean temperature within these two ROIs was considered as representative of the baseline temperature at the site of stimulation. The TFL was measured as the time between the onset of the thermal stimulus and the onset of the withdrawal response (visualized in the thermographic recording, with a 2-ms temporal resolution).

INTERACTIONS BETWEEN THERMOREGULATION AND NOCICEPTION 100

Power (%)

2) Spectral analysis. A FFT was used to estimate the frequency distribution of fluctuations during thermoneutral conditions. 3) Correlation analysis. Cross-correlations and Lissajous plots were used to study the linear correlation and lag between variables. The relative delay at which rBP occurred was used to estimate the time lag between two variables. Within the Lissajous plots, the dynamics of the coupling between variables was visualized by estimating trajectory direction at each time point by calculating the angle of the vector defined by the coordinates measured at that time point t and the coordinates measured at time point t ⫺ 5 s. 4) Statistical comparisons were done using the Mann-Whitney U-test. Significance level was set at P ⬍ 0.05.

2189

VLF domain

LF domain

Ttail-mid 50

Tcore

0 0.001

0.01

0.1

frequency (Hz)

RESULTS

Recordings in Awake Rats

[mean lag: 25 s (22–28 s); rBP ⫽ 0.97 (0.96 – 0.97)] and even more so, the fluctuations of Ttail-prox [mean lag: 69 s (63–76 s); rBP ⫽ 0.90 (0.88 – 0.91)]. Tcore fluctuations were delayed compared with fluctuations of Ttail-mid [mean lag: 211 s (98 –324 s); rBP ⫽ ⫺0.55 (⫺0.37 to ⫺0.73)]. Recordings in Anesthetized Rats A typical recording in an anesthetized rat maintained in thermoneutral condition. A typical example of a 3-h recording of a halothane-anesthetized rat maintained in thermoneutral conditions is shown in Fig. 4 (see also Supplemental Video S2). The thermographic pictures (Fig. 4A) are taken every minute from a 20-min sequence. Figure 4B shows the Tskin time courses measured at Tpaw-left and Tpaw-right at three sites of the tail (Ttail-prox, Ttail-mid, and Ttail-dist) and the environment (Tambient). Such as in the awake rat, maintained in a thermoneutral environment, Tskin showed marked and synchronous fluctuations at an approximate three-cycle/h (0.0007 Hz) frequency. Throughout the recording, the paws were, on average, warmer than the tail, and such as in the awake rats, the Ttail-prox was warmer than the Ttail-dist. The amplitude of the fluctuations was greater at the hind paws (⬃8°C) and the Ttail-dist (⬃7.5°C) compared with the Ttail-prox (⬃3.5°C). The speed of transitions from a vasoconstricted to a vasodilated state was faster than the speed of transitions from a vasodilated to a vasoconstricted state. There was a small but consistent temporal lag among the different temperature time courses, with changes occurring first in the hind paws, followed successively by changes in the distal and proximal parts of the tail. MAP and HR exhibited similar periodic fluctuations (Fig. 4C) in the 50-mmHg and 45-beats/min range, respectively. These fluctuations preceded Tskin changes in opposition of phase: vasodilatation and vasoconstriction were associated with drops and rises of MAP and HR, respectively. Throughout these 3 h of recording, the Tcore presented corresponding fluctuations, opposite to the skin variations (Fig. 4D). The amplitude variation reached 0.6°C during a single cycle, with



An illustrative example of a 3-h thermographic recording is shown in Fig. 2 (see also Supplemental Video S1, available in the data supplement online at the Journal of Neurophysiology Web site). The thermographic pictures (Fig. 2A) are taken every minute from a 20-min sequence. Figure 2B shows the temperature time courses of the skin measured at three sites of the tail (Ttail-prox, Ttail-mid, and Ttail-dist) and the environment (Tambient). Figure 2C shows the Tcore. During the 1st min of the recording, Tambient was ⬃26°C, and Tskin remained close to it, indicating a sustained vasoconstriction of the tail. During the following 100 min, Tambient increased progressively from 27°C to 28.5°C, and strong periodic ⌬Tskin was observed, with a frequency of approximately three to four cycles/h (0.0007– 0.0010 Hz). These periodicities of Tskin reflect fluctuations of the vasomotor tone, occurring when the animal is in a thermoneutral environment (Gordon 1990, 1993; Romanovsky et al. 2002). In the last 80 min of recording, Tambient increased further to 30°C, and the Tskin became more stable, close to Tcore, indicating a sustained state of vasodilation. Throughout the 3-h recording, the Tskin of the tail showed overall variations of up to 9°C and 7.5°C for the distal and proximal part of the tail, respectively. In contrast, the Tcore remained stable within a narrow range (37.2–37.6°C), with small, delayed fluctuations opposite the fluctuation of the temperature of the tail. The Tskin is dependent on both the vasomotor tone of the skin vessels and the Tambient and Tcore. The thermal exchanges can be normalized in terms of HLI, which when calculated for each ROI and time point [HLIROI ⫽ (TROI ⫺ Tambient)/(Tcore ⫺ Tambient); see MATERIALS AND METHODS], varies between zero (when the Tskin equals the Tambient) and one (when the Tskin equals the Tcore). Overall, recordings were performed with a mean Tambient of ⬃26°C and mean Tcore of 37.6°C (37.1– 38.1°C). When the tail remained in sustained, vasoconstricted or vasodilated states, HLI varied between 0 and 0.16 (⬃0 –2°C above Tambient) and between 0.71 and 0.85 (⬃1.8 –3.4°C below Tcore), respectively. In thermoneutral conditions with large fluctuations of vasomotor tone, ⌬HLI varied between 0.34 and 0.54 (⌬TROI ⬃3.4 –5.5°C). A spectral analysis of these signals showed that fluctuations of Tskin and Tcore occurred at a two- to seven-cycle/h (0.0006 – 0.0020 Hz) frequency (Fig. 3). A cross-correlation analysis was performed during the phases of fluctuating vasomotor tone. Within each recording, the time lag was estimated by finding the delay at which the rBP was found between the two signals. Across recordings, the fluctuations of Ttail-dist preceded the fluctuations of Ttail-mid

Fig. 3. Cumulative frequency histogram of the power spectrum of Tskin time courses, measured at the middle part of the tail (Ttail-mid) and Tcore in awake rats maintained in a thermoneutral condition. Note the relative concentration of power within the 0.0005- to 0.003-Hz range (equivalent to 2-7 cycles/h), with a main peak of activity at ⬃0.0009 – 0.001 Hz, corresponding to the large variations occurring at a rate of ⬃3.2–3.6 cycles/h. x-Axis: frequency in a logarithmic scale (Hz); y-axis: cumulative frequency histogram, expressed in percentage of experiments in which the vasomotor tone was fluctuating at a given frequency. For facilitating the comparisons with the MAP/HR variability literature, we colored the background for VLF and LF domains in orange and yellow, respectively (see Fig. 8).

2190

INTERACTIONS BETWEEN THERMOREGULATION AND NOCICEPTION laser-Doppler probe

A Tpaw-right

90

95

100

105

110

time (min)

Tpaw-left

Ttail-prox

35

Ttail-mid

°C

Ttail-dist

30

Tambient 25

wood

30

20

20

10

10

0

0

30 26 22

-10

-10

-20

-20

ΔHR (bpm)

30

34

Tcore (°C)

38.5

38

37.5

0

20

40

60

80

100

120

140

160

180

time (min)

Fig. 4. Individual example of a 3-h thermographic recording of the spontaneous variations of paws and tail temperature in a halothane-anesthetized rat in a thermoneutral condition. A: colored pictures are taken every minute from a 20-min sequence of a thermographic recording (indicated by the vertical gray background in B–D). B: Tskin time courses were obtained by averaging the temperatures within 5 ROIs located at Tpaw-left and Tpaw-right and Ttail-prox, Ttail-mid, and Ttail-dist. An additional ROI was located over a piece of wood, measuring the Tambient. C: MAP and HR were measured using a catheter inserted into the common carotid artery. MAP and HR are expressed as the difference from the mean of the entire recording [⌬MAP in mmHg (left ordinate); ⌬HR in beats/min (bpm) (right ordinate)]. Note the concomitant changes in MAP and HR in the direction opposite to the changes in Tskin. D: Tcore was measured using a rectal probe. The time course is synchronized with the triggering of HR rises (blue dotted arrows) and decreases (red dotted arrows). The thresholds for increasing and decreasing HR are shown as blue and red dots, respectively. These thresholds delimited an interthreshold range of ⬃0.3°C (dotted lines, gray background).

overall Tcore maintained within the 37.6 –38.4°C limits. The triggering of HR rises and decreases was synchronized with high and low values of Tcore. These thresholds delimited an interthreshold range of 0.3– 0.6°C. Overall, the MAP, HR, Tcore, and Tskin were highly correlated. Correlation between tail and paws Tskin. Across the different recordings, periods of sustained vasoconstriction, periods of sustained vasodilation, and periods characterized by large variations in vasomotor tone were observed, suggesting that minute changes in the temperature of the environment could induce profound and abrupt changes in the vasomotor state. Figure 5 summarizes the temporal relationship between the fluctuations in Tskin observed at the paws and tail throughout the 133 h of recording. The figure highlights the following three points. 1) The strong correlation and absence of temporal lag (P ⫽ 0.22) between the fluctuations of Tskin measured at the right and left paws (Fig. 5A). This was confirmed by the cross-correlation analysis [mean lag: 1 s (⫺3 to ⫹5 s); rBP ⫽ 0.90 (0.88 – 0.93)]. 2) The strong correlation between the fluctuations of Tskin measured at the paw and the tail when both are in a fluctuating state (Fig. 5B), with a relative delay of tail vs. paw fluctuations [mean lag of Tpaw-right vs. Ttail-mid: 60 s (50 – 69 s); rBP ⫽ 0.74 (0.71– 0.78)]. 3) The strong correla-

tion between fluctuations at the proximal and distal parts of the tail (Fig. 5C), with a relative delay of the proximal vs. distal parts of the tail [mean lag of Ttail-prox vs. Ttail-dist: 69 s (63–76 s); rBP ⫽ 0.90 (0.88 – 0.91)], similar to what was observed in the awake rats. A summary of the Tskin results obtained in all anesthetized rats is as follows. Overall, recordings were performed with a mean Tambient 24.1°C (23.9 –24.3°C) and mean Tcore 38.2°C (38.0 –38.3°C). When the tail remained in sustained vasoconstricted or vasodilated states, HLI varied between 0 and 0.19 (⬃0 –2.8°C above Tambient) and between 0.68 and 0.88 (⬃1.5– 3.1°C below Tcore), respectively. In thermoneutral conditions with large fluctuations of vasomotor tone, ⌬HLI varied between 0.24 and 0.47 (⌬TROI ⬃3.4 – 6.6°C). Such as in the awake rats, the amplitude of fluctuations increased over the length of the tail and was greatest at the distal part (mean ⌬T: 6.6°C) than the proximal part (mean ⌬T: 3.5°C). However, the greatest fluctuations were observed at the paws (mean ⌬T: 7.1°C). Correlation among Tskin, MAP, HR, and Tcore. In the thermoneutral conditions characterized by large, stationary fluctuations of paws and tail vasomotor tone, as illustrated in Fig. 4, we observed synchronous fluctuations of the MAP and HR.



D

temperature (°C)

C

Tpaw-left Tpaw-right Ttail-prox Ttail-mid Ttail-dist Tambient

38

ΔMAP (mmHg)

B


1

1

HLIpaw-left

HLIpaw-left

A

0.5

0.5

0

0 0

0.5

0

1

HLIpaw-right

0.5

0.5

0 0

0.5

1

0

HLIpaw-right

0.5

1

HLIpaw-right

1

1

HLItail-prox

HLItail-prox

1

1

HLItail-mid

1

0

C

0.5

HLIpaw-right

0.5

0

Fig. 5. Lissajous curves describing the temporal relationship between the Tskin measured in all of the recordings performed in anesthetized rats in a thermoneutral condition. Left: all measures superimposed using a gray scale, illustrating the density of points (number of superimposed points); right: all measures superimposed but using a color scale coding for the direction of the change in each of the 2 compared variables. A: left and right paws (HLIpaw-right vs. HLIpaw-left). In most cases, the overall temperature variations of the right and left hind paws were highly correlated (diagonal orientations). On rare occasions, a dissociation was observed between the 2 variables (vertical and horizontal orientations). B. paw and tail (HLIpaw-right vs. HLItail-mid). In 35% of the cases, synchronous fluctuations of the paw and tail were observed (diagonal orientations). In 25% of the cases, sustained vasoconstriction of the paw was observed, whereas the tail continued to fluctuate (vertical orientations in the left part of the graph). In 28% of the cases, sustained vasodilation of the tail was observed, whereas the paw continued to fluctuate (horizontal orientations in the upper part of the graph). C: Proximal and distal parts of the tail (HLItail-dist vs. HLItail-prox). There was a strong correlation between the fluctuations observed at the HLItail-prox and HLItail-dist. HLItail-prox was always warmer than HLItail-dist (plots above the bisecting line).

0.5

50

density

direction

0

0 0

0.5

1

0

HLItail-dist

0.5

1

HLItail-dist

Grand mean fluctuations of MAP and HR were 46 mmHg (43–50 mmHg) and 45 beats/min (40 –50 beats/min), respectively. Drops in MAP and HR were associated with paw and tail vasodilation, whereas rises in MAP and HR were associated with paw and tail vasoconstriction. As determined by cross-correlation analyses, the changes in MAP shortly preceded the changes in HR, which themselves preceded the changes in Tskin {lag MAP-HR: 32 s (25– 41 s), rBP ⫽ 0.78 (0.74 – 0.82); lag MAP-Tpaw-right: 184 s (164 –203 s), rBP ⫽ ⫺[0.62 (0.58 – 0.67)]; lag MAP-Ttail-mid: 310 s (288 –333 s), rBP ⫽ ⫺[0.59 (0.55– 0.64)]}.

The Tcore also exhibited small fluctuations [grand mean ⌬T: 0.7°C (0.6 – 0.8°C)]. These fluctuations were in the opposite direction of the fluctuation of the temperature of the tail and paws {Ttail-mid-Tcore: 32 s (20 – 45 s), rBP ⫽ ⫺[0.71 (0.67– 0.75)]; Tpaw-right-Tcore: 140 s (167–113 s), rBP ⫽ ⫺[0.64 (0.60 – 0.69)]}. Drops in Tcore were associated with paws and tail vasodilation, whereas rises in Tcore were associated with paws and tail vasoconstriction. Overall, the changes of Tcore were preceded by ⌬MAP and ⌬HR, 5– 6 min earlier, as determined by cross-correlation analyses [MAP-Tcore: 354 s (321–384 s), rBP ⫽ 0.54 (0.50 – 0.59); HR-Tcore: 271 s (238 –



HLItail-mid

B

2191

2192


303 s), rBP ⫽ 0.63 (0.59 – 0.68)]. Figure 6 shows the relationship between ⌬HR and ⌬Tcore (Fig. 6A) and between ⌬HR and lag-realigned ⌬Tcore (Fig. 6B). The individual linear correlations were significant in nine of 36 and 27 of 36 cases, respectively. Overall, these effects were not (F1–3,858 ⫽ 0.31; P ⫽ 0.58) or were highly (F1–3,584 ⫽ 1,444; P ⬍ 0.0001) significant, respectively. Interestingly, the Tcore increased by a mean 0.38°C (0.26 – 0.51°C) when HR increased by 50 beats/min. If one considers the periods of sustained, stationary vasomotor tone observed during the overall 133 h of recording, one

A

0.5

source of df sum of mean variation squares square ΔTc rat rat*ΔTc

55.2

F

P

55,2 0.313

0.58

35 13.1 0.376 0.002 > 0.99 35 178900 5111.4 28.98 < 0.0001 3858 680393

176.4

0

ΔTc = [0.0041 (0.014-0.067)]*ΔHR; R2 = 0.08

-0.5

B

0.5

source of df sum of mean squares square variation ΔTc rat rat*ΔTc

lag-realigned ΔTcore (°C)

error

F

P

1 154479 154479 1444 < 0.0001 35 5.98 0.17 0.002 > 0.99 35 138936 3969.6 37.2 < 0.0001 3584 383404 106.97

Correlation between TFL and Tskin: a Behavioral Study 0

-0.5 -75

ΔTc = [0.0077 (0.0052-0.0102)]*ΔHR; R2 = 0.39

0

75

ΔHR (bpm) Fig. 6. Correlation of ⌬HR with ⌬Tcore in anesthetized rats in thermoneutral condition: overall results. A: in each individual case, the linear regression between ⌬HR and ⌬Tcore was calculated. For clarity of presentation, only the regression lines are drawn (n ⫽ 36; black lines). Abscissa: ⌬HR (beats/min); ordinate: ⌬Tcore (°C). In 9 of 36 cases, HR covaried with Tcore, but overall, these effects were not significant (grand mean shown as dotted white line). B: in each individual case, the rBP between the ⌬Tcore and ⌬HR was calculated. The relative delay at which rBP occurred was used to estimate the lag between HR and Tcore. Then, linear regression calculated between ⌬HR and lagrealigned ⌬Tcore was built as in A. The individual linear correlations became significant in 27 of 36 cases (black lines). Overall, HR covaried with Tcore, recorded 4 –5 min later (dotted white line). The effect sizes [R2/(1 ⫺ R2)] were 0.09 (small) and 0.63 (large) in A and B, respectively. df, degrees of freedom; F, F-test; P, P value.

As a final complement to the experiments described above, we performed a series of measurements of the TFL in conditions of thermoneutrality. The tail-flick test remains one of the most widely used in rodents for the study of pain: a constant thermal radiation is applied to the tail that elicits a withdrawal. The reaction time is recorded, and its variations are interpreted as inverse variations of pain sensitivity. The aim was to evaluate the impact of the ⌬Tskin on a variable commonly interpreted as a “pain index”. The individual recording in Fig. 10A shows the clear inverse relationship between tail temperature and TFL: lower Tskin was associated with longer reaction times, whereas higher Tskin was associated with shorter reaction times. In spite of large interindividual variability in TFL, this correlation was significant across the eight animals (Fig. 10B). Overall, the mean extent of the ⌬Tskin and TFL was 4.0°C (3.0 – 4.9°C) and 1.8 s (1.4 –2.2 s), respectively. DISCUSSION

The thermographic recordings in awake and halothane-anesthetized rats show that when rats are maintained in a thermoneutral environment, the vasomotor tone of the paws and tail is



ΔTcore (°C)

error

1

sees that the large variations described above declined with the loss of balance toward vasoconstriction or vasodilatation. This is summarized by the consideration of the mean HLItail-mid—as index of vasomotor tone—and the interquartile ranges of MAP (Fig. 7A) and HR (Fig. 7B)—as respective indexes of variability. Interestingly, the mean MAP and HR (Fig. 7C) were not significantly affected by the vasomotor tone. Summary: temporal aspects. Spectral analysis of the signals showed that these fluctuations occurred with a frequency of two to seven cycles/h (0.0006 – 0.0020 Hz; Fig. 8). In summary, the fluctuations of MAP, HR, Tskin, and Tcore were highly correlated and followed a chronological order, as changes successively occurred at the following: 1) MAP, 2) HR, 3) the hind paws, 4) the tail, and 5) Tcore (Fig. 9A). Figure 9B provides a schematic summary of the chronologic order and amplitude of changes. When a relatively high level of Tcore was reached, the MAP decreased suddenly, followed within 30 s by a decrease in HR. The vasodilation of the tail followed invariably within ⬃5 min, often associated with vasodilation of the hind paws, slightly preceding the vasodilation of the tail by ⬃2 min. The outcome of these changes was a decrease of the Tcore, ⬃6 min after the onset of the ⌬MAP and ⌬HR. Conversely, when the Tcore was decreased by a few tenths of degrees, sympathetic activation elicited the opposite variations, leading to an increase in Tcore. Overall, a single cycle successively comprised the following: 1) a short period of sympathetic deceleration, followed by a plateau often longer, although more variable than the initial parts of the pattern (red hatched areas in Fig. 9B), and 2) an ⬃7- to 8-min period of sympathetic activation, followed by a plateau of the same order of duration, also more variable (blue hatched areas in Fig. 9B). The amplitudes of these changes were not trivial: ⌬MAP in the ⬃40- to 50-mmHg range, ⌬HR in the 40 –50 beats/min range, temperature in the 3– 6°C range for the dorsal facet of the tail (proximal ⬍ distal), and ⬃7°C for the sole of the paws. This resulted in an ⬃0.7°C ⌬Tcore.


A

100

VLF domain

Power (%)

characterized by abrupt, large-amplitude, and synchronous transitions between vasoconstriction and vasodilation, occurring at a mean rate of three to four cycles/h. These changes in vasomotor tone were synchronous with significant changes in MAP and HR. Cross-correlation analysis of the concomitantly recorded signals revealed that these changes followed a chronological order: 1) an increase in MAP, followed by 2) an

2193

LF domain

Tpaw-right Ttail-mid

50

Tcore ΔMAP

MAP interquartile range

ΔHR

50

0

mmHg

0.001

0.01

0.1

frequency (Hz)

25

0

B

HR interquartile range

50

bpm

increase in HR within 30 s and 3) vasoconstriction of the tail and paws within 3–5 min; the outcome is an increase in Tcore within 6 min. Conversely, a sympathetic inhibition led to the following: 1) a decrease in MAP, followed by 2) a decrease in HR and 3) vasodilation of the tail and paws; the outcome is a decrease in Tcore. Most importantly, we found that the large changes in Tskin, induced by these thermoregulatory mechanisms, had a strong incidence on the latency in the tail-flick test, one of the most-used behavioral response to “assess pain” in rodents: mean variations in tail temperature of ⬃1°C led to significant inverse variations in reaction times of ⬃0.5 s.

25

Variability and Synchrony of Tail and Paw Vasomotor Tone in Thermoneutral Conditions

0 0

C

0.5

1

mean ABP and HR 500

bpm

mmHg

200

250

100

0

0 0

0.5

mean HLItail-mid

1

In rats, it is well known that the tail plays an important role in thermoregulation, acting as a radiator. O’Leary et al. (1985) showed that the vasculature of the tail is not endowed with an active vasodilator system. Instead, control of blood flow through the arterial-venous anastomoses of the tail is mediated exclusively through changes in sympathetic constrictor tone (Grant 1963; Johnson and Gilbey 1994, 1996, 1998; Richard-

Fig. 7. Relationship between vasomotor tone and MAP/HR variability in anesthetized rats. The sustained, stationary vasomotor-tone sequences observed during at least 40 min were sampled from the overall 133 h of recording. For clarity of presentation, the mean HLItail-mid was considered as representative of the vasomotor status of the animals (123 individual stable sequences) and was regrouped in 0.1 step classes (abscissa). A: interquartile ranges of MAP. B: interquartile ranges of HR. C: mean MAP (left ordinate) and HR (right ordinate). Whereas the mean MAP and HR were not significantly affected by the vasomotor tone, their variability declined with the loss of HLI balance toward vasoconstriction or vasodilatation. The mean Tambient and Tcore were 24.1°C (23.9 –24.3°C) and 38.2°C (38.0 –38.3°C), respectively. ABP, arterial blood pressure.



Fig. 8. Cumulative frequency histogram of the power spectrum of results concerning time courses of skin Tpaw-right and Ttail-mid, ⌬Tcore, ⌬MAP, and ⌬HR in anesthetized rats in a thermoneutral condition, in which the vasomotor tone was fluctuating. Abscissa: logarithmic scale of frequencies in hertz; ordinate: cumulative frequency histogram expressed in percentage of results. Note the relative concentration of power within the 0.0005- to 0.003-Hz range (equivalent to 2 and 7 cycles/h) with a main peak of activity ⬃0.001– 0.002 Hz, corresponding to the large variations occurring at a rate of 3–7 cycles/h. Note an eminent, additional peak frequency at 0.17 Hz observed for MAP and HR, together with others in the 0.2- to 0.5-Hz range, referred to as LF in the MAP/HR variability literature (Julien et al. 2008; Leung and Mason 1996; Persson et al. 1992; Stauss 2007). For facilitating the comparisons with this literature, we colored the background for VLF and LF domains in orange and yellow, respectively.

2194


A

p < 0.05 p < 0.01

time lag with MAP HR

T

core

MAP 0

B

Ttail-mid

Tpaw-right 1

2

3

4

5

6

min

7

summary sympathetic activation

sympathetic inhibition

sympathetic activation

upper set point

Tcore lower set point ~ 6 min

~ 6 min

MAP HR ~ 7 min

Tpaw-right Ttail-mid ~ 3-5 min

~ 3-5 min

Fig. 9. A: graphic representation of the time lags between MAP and the other variables, namely HR, Tpaw-right, Ttail-mid, and Tcore. B: schematic summary of results. The relative threshold Tcore set points for inhibiting and activating the sympathetic system are shown as red and blue dots, respectively. Sympathetic inhibition (red area) causes a decrease of MAP and HR associated with skin vasodilatation, itself leading to an increase in blood flow and Tskin at both the level of the tail and the hind paws, within ⬃3 and ⬃5 min, respectively. The outcome is an increase in heat loss that decreases the Tcore within ⬃6 min. When the sympathetic activation switched on (blue area), the opposite variations occur: increase of MAP and HR associated with skin vasoconstriction, itself leading to a decrease in blood flow and Tskin. The outcome is a decrease in heat loss that raises the Tcore within ⬃6 min. The hatched areas represent the variable periods of the pattern.

son et al. 1991). In contrast, it is often considered that the paws are not involved in thermoregulation, although several previous studies have shown the contrary (see INTRODUCTION). Here, we found that rats maintained in a thermoneutral state exhibit large and synchronous paw and tail ⌬Tskin, indicating that both are clearly involved in thermoregulation. In fact, paw ⌬Tskin were, on average, of greater magnitude than tail ⌬Tskin. Considering that the surface of the rat feet and tail represent ⬃10% and ⬃7% of the total body surface, respectively (Lin et al. 1979), our results suggest that the paws may actually play a greater role in thermoregulation than the tail. In both awake and anesthetized rats, ⌬Tskin occurred at a mean frequency of approximately three to four cycles/h, compatible with previous reports by Young and Dawson (1982). However, several previous studies did not report such fluctuations. For example, in unrestrained and undisturbed rats, Gordon et al. (2002) only reported cyclic ⌬Tskin in the range of one to two cycles/h. These slower variations were most probably related to BRAC, which are characterized by behavioral activities, such as eating, occurring with periods in the 1- to 2-h

Synchrony of Tail and Paw Vasomotion with ⌬MAP and ⌬HR The variations in tail and paws vasomotion observed in thermoneutral conditions were associated with synchronous



~ 4.5 min

range, and are accompanied by increases in MAP, HR, and BAT activity (Blessing et al. 2012; Closa et al. 1993; HolsteinRathlou et al. 1995; Ootsuka et al. 2009). That we are dealing with two different phenomena is well illustrated by the opposite phenomenological relationship between cardiovascular and vasomotor events: in BRAC, the increases of MAP and HR are followed by vasodilation of the tail, whereas in our study, they were followed by vasoconstriction. In anesthetized rats, the heat production by muscular activity (Girardier et al. 1995; Morrison 1968) is blocked, whereas the activity of BAT is strongly reduced (Dicker et al. 1995; Ohlson et al. 2003), thus favoring the focus of our study on a short-term regulation. In contrast, it is very likely that the effect of BRAC and other motor activities in unrestrained rats blurs the observation of the faster cycles described here, possibly because the periods of a stable thermoneutral state are too short. The ⌬Tskin, occurring in the left and right paws, was generally coupled. That is, transitions from vasodilated to vasoconstricted states and the reverse occurred at the same time in the two paws. The ⌬Tskin of the paws and tail was also highly synchronous. However, in several instances, vasodilation of the tail was observed without concomitant vasodilation of the paws, suggesting that the thermoregulatory mechanisms responsible for these changes are able to control independently the vasomotor tone of the paws and tail. The medial region of the preoptic area of the hypothalamus controls, for thermoregulatory purposes, both the vasomotor tone (Morrison and Nakamura 2011; Nakamura 2011) and the HR (Cao et al. 2004; Cao and Morrison 2006; DiMicco and Zaretsky 2007; Nakamura et al. 2004), through projections to the rostral medullary raphe. This most likely explains our observation of largely synchronous fluctuations in HR and vasomotor tone in thermoneutral conditions. That vasodilation of the tail and paws can occur independently suggests that these efferent structures have the ability to generate targetspecific responses. Such output specificity has already been suggested in the rostral ventrolateral medulla, in which sympathoexcitatory cells could be organized into functional subgroups controlling the heart or specific vascular beds (Jänig 2006; Morrison 2001). The ⌬Tskin consisted of a series of relatively abrupt transitions between temperatures close to Tambient and temperatures close to Tcore, indicating that the thermoregulatory mechanisms operate in an on-off fashion, in which the tail and paws are either in a state of complete vasoconstriction or a state of complete vasodilation. On average, the amplitude of the changes in Tskin was greater at the paws compared with the tail and greater at the distal part of the tail compared with the proximal part of the tail. These differences are probably related to differences in vasculature. For example, Wu et al. (1995) found that as one runs toward the tail tip, the diameter of the ventral artery remains large, and a very large number of arterial-venous anastomoses are seen, suggesting a higher blood perfusion of the distal part of the tail compared with the proximal part of the tail.


A

B

30

5.5

6

5

5

4.5 28 4 26

tail-flick latency (s)

temperature of the tail (°C)

32

2195

Source of Sum of Mean variation df squares square F Significance 1 90.9 TFL 90.9 115.5 < 0.0001 7 29.7 5.4 < 0.0001 4.2 rat 7 33.3 6.0 < 0.0001 rat *TFL 4757 Error 260 204.6 0.787

4

3

3.5

2

3

1

TFL = -1.78*Ttail + 8.4

24 0

20

40

60

time (min)

R2 = 0.22; p < 0.0001

25

27

29

Ttail (°C)

31

33

changes in MAP and HR. Specifically, states of vasoconstriction were associated with increases in MAP and HR, whereas states of vasodilation were associated with decreases in MAP and HR. Spectral analyses of these recordings showed that the fluctuations observed in all signals peaked at 0.001– 0.002 Hz, both in awake and anesthetized rats. This is in accordance with earlier observations regarding the patterns of ⌬Tskin of the tail (Young and Dawson 1982) and ⌬MAP or ⌬HR (Julien et al. 2008; Leung and Mason 1996; Persson et al. 1992; Stauss 2007). Few articles have focused on thermoregulation patterns below circadian rhythm, including BRAC, in the so-called VLF (⬍0.04 – 0.2 Hz), recognized the MAP/HR variability literature. Note that in this respect, the recording of LF waves (0.2– 0.8 Hz), including Mayer waves (Chapuis et al. 2004; Julien 2006; Stauss 2007), was not our objective. Our data collection was made at a 1-Hz rate, excluding the investigation of frequencies ⬎0.5 Hz. It follows that we were not able to detect all Mayer waves. However, several peaks described already were verified in our recordings (e.g., 0.17 Hz) but did not concern thermal exchanges submitted to inertia. Note that the autonomous basic sympathetic rhythm (0.4 –1.2 Hz) was identified in the activity of single sympathetic postganglionic neurons innervating the caudal ventral artery of the rat tail (Johnson and Gilbey 1996). Interestingly, this sympathetic vasoconstrictor activity is preserved following collicular decerebration, suggesting that the oscillator network underlying the basal rhythm generation is located in the brain stem (Collins and Gilbey 2003). We also observed a significant, positive correlation between Tcore and HR: increases in HR were associated with increases in Tcore, whereas decreases in HR were associated with decreases in Tcore. Similar observations have been observed at the scale of circadian rhythm (Gordon 1994) and related to locomotion (Büttner and Wollnik 1982; Ootsuka et al. 2009). Note that the correlation between HR and Tcore, as contemporaneous variables, was weak but increased very much when these

variables were postponed by ⬃5 min. Since the only active muscle in our experiments in anesthetized animals was the heart, such a close correlation suggests that the temperature regulation was achieved by both controlling heat gain through heartbeat-induced thermogenesis and heat loss through cutaneous vasoconstriction. Thermoneutrality Is Metastable Many definitions have been proposed for thermoneutrality. The classical definition refers to a minimal and constant basal metabolic rate (Gordon 1993). The Commission for Thermal Physiology of the International Union of Physiological Sciences (2001) defined the thermoneutral zone as “the range of ambient temperature at which temperature regulation is achieved only by control of sensible heat loss, i.e., without regulatory changes in metabolic heat production or evaporative heat loss.” To achieve and maintain thermoneutrality, the trunk of the animal was maintained at Twarm, slightly above Tcore. By considering blood-pressure variations as the premonitory observable sign of vasomotor changes in the tail and paws, we found that vasoconstriction and vasodilation were triggered at a mean Tcore of 38.0°C and 38.3°C, respectively. The narrowness of this range is in line with the results of several studies in humans and rats. For example, the mean rectal threshold temperature for eliciting responses of sympathetic fibers in the ventral tail artery in anesthetized rats was reported in the 38.1–38.3°C range (Ootsuka and McAllen 2006; Owens et al. 2002; Tanaka et al. 2007). Interestingly, the basic sympathetic vasoconstrictor activity becomes active when Tcore falls below ⬃38°C in anesthetized rats (Johnson and Gilbey 1998). Similarly, the interthreshold range in humans, as defined by sweating and vasoconstriction thresholds, is usually only 0.2– 0.4°C (Lopez et al. 1994). It is unlikely that metabolic thermogenesis was involved in the cyclic ⌬Tcore that we observed. Indeed, the neurons regu-



Fig. 10. Fluctuation of the TFL during cyclic changes of the tail Tskin in rats held in the conditions of thermoneutrality described above. The reaction time of the responses to radiant heat stimuli applied to the middle of the tail was measured, while the thermographic camera monitored the Tskin. The stimulus was a 6-mm-diameter circular spot, elicited by a laser stimulator device that includes a collimated measurement of Tskin and accurate feedback monitoring to achieve a constant rate of heating (6°C/s). A: individual example. The blue line represents the mean temperature (ordinate on the left of the graph), measured at 2 locations of the tail, proximal and distal to the stimulated target. The reaction time is shown as vertical red bars with ordinate on the right of the graph. The individual recording shows the clear inverse relationship between tail temperature and TFL: lower Tskin was associated with longer reaction times, whereas higher Tskin was associated with shorter reaction times. B: overall results expressed in terms of the relationship between the basal Tskin and the reaction time to the stimulus. In spite of an interindividual variability, the general trend was very similar.

2196


lating tail fiber activity are much more sensitive and respond earlier to changes in Tcore than those regulating BAT (McAllen et al. 2010; Ootsuka and McAllen 2006; Owens et al. 2002). We have already mentioned that halothane anesthesia reduces the activity of the BAT (Dicker et al. 1995; Ohlson et al. 2003). Most importantly, from the perspective of paw and tail vasomotor tone and temperature, the thermoneutral condition is a highly unstable condition. Indeed, during this thermoneutrality, Tcore is essentially maintained by adjusting heat loss through large on-off transitions of skin vasomotor tone, bringing Tskin from temperatures close to Tambient (complete vasoconstriction) to temperatures close to Tcore (complete vasodilation). Implications for Pain Studies

Conclusion In the present study, we show that the temperature of the paws and tail of awake and anesthetized rats maintained in thermal neutrality is characterized by cyclic, large-amplitude changes in temperature. These changes were synchronous with changes in MAP and HR and contributed to maintaining the Tcore within a narrow range. Most importantly, we found that these thermoregulatory changes in Tskin had a profound effect on the latency of behavioral responses commonly used to study nociception, thus questioning their validity to study the physiological mechanisms of acute and chronic pain. ACKNOWLEDGMENTS We thank Professor Pierre Le Bars for advice on the frequency domain analysis and Professors François Cesselin and Léon Plaghki for advice toward the preparation of the manuscript. GRANTS Support for N. El Bitar was provided by a grant from the Société Française d’Etude et de Traitement de la Douleur (SFETD) et l’Institut UPSA de la douleur (IUD). DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the authors. AUTHOR CONTRIBUTIONS Author contributions: N.E.B., B.P., and D.L.B. conception and design of research; N.E.B., B.P., E.K., and I.P. performed experiments; N.E.B., B.P., E.K., I.P., A.M., and D.L.B. analyzed data; N.E.B., B.P., E.K., I.P., and D.L.B. interpreted results of experiments; N.E.B., A.M., and D.L.B. prepared figures; N.E.B. and D.L.B. drafted manuscript; N.E.B., B.P., A.M., and D.L.B. edited and revised manuscript; N.E.B., B.P., E.K., I.P., A.M., and D.L.B. approved final version of manuscript. REFERENCES Aukland K, Wiig H. Hemodynamics and interstitial fluid pressure in the rat tail. Am J Physiol Heart Circ Physiol 247: H80 –H87, 1984. Benoist JM, Pincedé I, Ballantyne K, Plaghki L, Le Bars D. Peripheral and central determinants of a nociceptive reaction: an approach to psychophysics in the rat. PLoS One 3: e3125, 2008. Blessing W, Mohammed M, Ootsuka Y. Heating and eating: brown adipose tissue thermogenesis precedes food ingestion as part of the ultradian basic rest-activity cycle in rats. Physiol Behav 105: 966 –974, 2012. Büttner D, Wollnik F. Spontaneous short-term fluctuations in the daily pattern of heart rate, body temperature and locomotor activity in the laboratory rat. Lab Anim 16: 319 –326, 1982. Cao WH, Fan W, Morrison SF. Medullary pathways mediating specific sympathetic responses to activation of dorsomedial hypothalamus. Neuroscience 126: 229 –240, 2004.



To characterize the effect of these thermoregulatory vasomotor variations on the behavioral responses to thermonociceptive stimuli, we examined the relationship between the latency of the tail-flick test and the changes in tail Tskin associated with the thermoneutral state. We chose the tail-flick test (d’Amour and Smith 1941), as it is emblematic and most importantly, highly representative of a series of frequently used tests based on the measurement of a reaction time to heating the extremities of the animal body, including the “paw withdrawal test” (Hargreaves et al. 1988) and the “hot-plate test” (Woolfe and MacDonald 1944). We found that the latency of the tail-flick test is strongly dependent on the temperature of the tail at the moment of stimulation. Lower temperatures translated into an increase in withdrawal latency, whereas higher temperatures translated into a decrease in withdrawal latency. These differences in latency (which would have been interpreted as a sign of hypoalgesia or hyperalgesia) can be explained easily by the fact that the response to a given stimulus is not only dependent on the physical properties of the stimulus but also on the biophysical properties of the stimulated skin, such as its temperature, as well as the biophysical properties of the nerve fibers conducting the nociceptive input, such as their temperature, which is known to affect conduction velocity markedly. Regardless of the mechanisms explaining the effect of tail temperature on the latency of the tail-flick test, our results indicate that one should avoid recording such responses in conditions of thermoneutrality, as the large variations in vasomotor tone and ⌬Tskin associated with this state are likely to be an important source of variability in the latency of the elicited responses, thus limiting their usefulness as a surrogate measure of pain perception. In support of our view, previous studies have already suggested that spontaneous hypertension or the peripheral administration of pressor agents increases the latency of the tail-flick and hot-plate tests (Ghione 1996; Maixner 1991; Randich and Maixner 1984; Zamir and Maixner 1986). On this account, it is often believed that stimulation of the paw should be preferred to stimulation of the tail, because the paw would have the advantage of not being involved in thermoregulation. The results of the present study, showing that the changes in paw temperature are of greater amplitude than the changes of tail temperature, demonstrate that this is definitely not true, as already suggested by previous reports (Gordon and Watkinson 1988; Grant 1963; Kanosue et al.

1994; Key and Wigfield 1992, 1994; Lin et al. 1978, 1979; Thompson and Stevenson 1965; Zhang et al. 1995). Finally, our results indicate that when interpreting the effect of a given experimental or pharmacological manipulation on the latency of these responses, one should always consider the possibility that the observed effect is not due to a modulation of the state of the nociceptive system (e.g., a lesion thought to induce sensitization of the nociceptive system or a pharmacological compound thought to induce analgesia). Indeed, the observed changes in response latencies could be explained by an effect of the manipulation on the thermoregulatory controls.


Johnson CD, Gilbey MP. On the dominant rhythm in the discharges of single postganglionic sympathetic neurones innervating the rat tail artery. J Physiol 497: 241–259, 1996. Johnson CD, Gilbey MP. Sympathetic activity recorded from the rat caudal ventral artery in vivo. J Physiol 476: 437– 442, 1994. Julien C. The enigma of Mayer waves: facts and models. Cardiovasc Res 70: 12–21, 2006. Julien C, Saul JP, Parati G. Very low frequency blood pressure fluctuations: not only myogenic responsiveness. J Hypertens 26: 1065–1068, 2008. Kanosue K, Yanase-Fujiwara M, Hosono T. Hypothalamic network for thermoregulatory vasomotor control. Am J Physiol Regul Integr Comp Physiol 267: R283–R288, 1994. Key BJ, Wigfield CC. Changes in the tail surface temperature of the rat following injection of 5-hydroxytryptamine into the ventrolateral medulla. Neuropharmacology 31: 717–723, 1992. Key BJ, Wigfield CC. The influence of the ventrolateral medulla on thermoregulatory circulations in the rat. J Auton Nerv Syst 48: 79 – 89, 1994. Knoppers AT. La queue du rat, témoin de la régulation thermique. Arch Néer Physiol 26: 364 – 406, 1942. Le Bars D, Gozariu M, Cadden SW. Animal models of nociception. Pharmacol Rev 53: 597– 652, 2001. Le Bars D, Hansson P, Plaghki L. Current animal test and models of pain. In: Pharmacology of Pain, edited by Beaulieu P, Lussier D, Porreca F, and Dickenson AH. Seattle: IASP, 2009, p. 475–504. Leung CG, Mason P. Spectral analysis of arterial blood pressure and raphe magnus neuronal activity in anesthetized rats. Am J Physiol Regul Integr Comp Physiol 271: R483–R489, 1996. Lin MT, Chen CF, Pang IH. Effect of ketamine on thermoregulation in rats. Can J Physiol Pharmacol 56: 963–967, 1978. Lin MT, Chern YF, Liu GG, Chang TC. Studies on thermoregulation in the rat. Proc Natl Sci Counc ROC 3: 46 –52, 1979. Little RA, Stoner HB. The measurement of heat loss from the rat’s tail. Q J Exp Physiol Cogn Med Sci 53: 76 – 83, 1968. Lomax P. Measurement of ‘core’ temperature in the rat. Nature 210: 854 – 855, 1966. Lopez M, Sessler DI, Walter K, Emerick T, Ozaki M. Rate and gender dependence of the sweating, vasoconstriction, and shivering thresholds in humans. Anesthesiology 80: 780 –788, 1994. Maixner W. Interactions between cardiovascular and pain modulatory systems: physiological and pathophysiological implications. J Cardiovasc Electrophysiol 2: s3–s12, 1991. McAllen RM, Tanaka M, Ootsuka Y, McKinley MJ. Multiple thermoregulatory effectors with independent central controls. Eur J Appl Physiol 109: 27–33, 2010. Morrison SD. The constancy of the energy expended by rats on spontaneous activity, and the distribution of activity between feeding and non-feeding. J Physiol 197: 305–323, 1968. Morrison SF. Differential control of sympathetic outflow. Am J Physiol Regul Integr Comp Physiol 281: R683–R698, 2001. Morrison SF, Nakamura K. Central neural pathways for thermoregulation. Front Biosci 16: 74 –104, 2011. Nakajima Y, Nose H, Takamata A. Comparison between tail skin blood flow measurements by ultrasonic Doppler flowmetry and plethysmography during heating in anesthetized rats. Jpn J Physiol 49: 121–124, 1999. Nakamura K. Central circuitries for body temperature regulation and fever. Am J Physiol Regul Integr Comp Physiol 301: R1207–R1228, 2011. Nakamura K, Matsumura K, Hubschle T, Nakamura Y, Hioki H, Fujiyama F, Boldogkoi Z, Konig M, Thiel HJ, Gerstberger R, Kobayashi S, Kaneko T. Identification of sympathetic premotor neurons in medullary raphe regions mediating fever and other thermoregulatory functions. J Neurosci 24: 5370 –5380, 2004. O’Leary DS, Johnson JM, Taylor WF. Mode of neural control mediating rat tail vasodilation during heating. J Appl Physiol 59: 1533–1538, 1985. Ohlson KB, Lindahl SG, Cannon B, Nedergaard J. Thermogenesis inhibition in brown adipocytes is a specific property of volatile anesthetics. Anesthesiology 98: 437– 448, 2003. Ootsuka Y, de Menezes RC, Zaretsky DV, Alimoradian A, Hunt J, Stefanidis A, Oldfield BJ, Blessing WW. Brown adipose tissue thermogenesis heats brain and body as part of the brain-coordinated ultradian basic rest-activity cycle. Neuroscience 164: 849 – 861, 2009. Ootsuka Y, McAllen RM. Comparison between two rat sympathetic pathways activated in cold defense. Am J Physiol Regul Integr Comp Physiol 291: R589 –R595, 2006.



Cao WH, Morrison SF. Glutamate receptors in the raphe pallidus mediate brown adipose tissue thermogenesis evoked by activation of dorsomedial hypothalamic neurons. Neuropharmacology 51: 426 – 437, 2006. Carrive P, Churyukanov M, Le Bars D. A reassessment of stress-induced “analgesia” in the rat using an unbiased method. Pain 152: 676 – 686, 2011. Chapuis B, Vidal-Petiot E, Orea V, Barrès C, Julien C. Linear modelling analysis of baroreflex control of arterial pressure variability in rats. J Physiol 559: 639 – 649, 2004. Closa D, Gómez-Sierra JM, Latres E, Alemany M, Remesar X. Short-term oscillations of aortic core body temperature and thermogenic organ blood flow in the rat. Exp Physiol 78: 243–253, 1993. Collins DR, Gilbey MP. Cutaneous sympathetic motor rhythms in the decerebrate rat. Neuroscience 117: 981–989, 2003. Covino BG, Dubner R, Gybels J, Kosterlitz HW, Liebeskind JC, Sternbach RA, Vycklický L, Yamamura H, Zimmermann M. Ethical standards for investigations of experimental pain in animals. Pain 9: 141–143, 1980. d’Amour FE, Smith DL. A method for determining loss of pain sensations. J Pharmacol Exp Ther 72: 74 –79, 1941. Dawson NJ, Keber AW. Physiology of heat loss from an extremity: the tail of the rat. Clin Exp Pharmacol Physiol 6: 69 – 80, 1979. Dicker A, Ohlson KB, Johnson L, Cannon B, Lindahl SG, Nedergaard J. Halothane selectively inhibits nonshivering thermogenesis. Possible implications for thermoregulation during anesthesia of infants. Anesthesiology 82: 491–501, 1995. DiMicco JA, Zaretsky DV. The dorsomedial hypothalamus: a new player in thermoregulation. Am J Physiol Regul Integr Comp Physiol 292: R47–R63, 2007. Gemmell RT, Hales JR. Cutaneous arteriovenous anastomoses present in the tail but absent from the ear of the rat. J Anat 124: 355–358, 1977. Ghione S. Hypertension-associated hypalgesia. Evidence in experimental animals and humans, pathophysiological mechanisms, and potential clinical consequences. Hypertension 28: 494 –504, 1996. Girardier L, Clark MG, Seydoux J. Thermogenesis associated with spontaneous activity: an important component of thermoregulatory needs in rats. J Physiol 488: 779 –787, 1995. Gordon CJ. 24-Hour control of body temperature in rats. I. Integration of behavioral and autonomic effectors. Am J Physiol Regul Integr Comp Physiol 267: R71–R77, 1994. Gordon CJ. Temperature Regulation in Laboratory Rodents. New York: Cambridge University Press, 1993. Gordon CJ. Thermal biology of the laboratory rat. Physiol Behav 47: 963– 991, 1990. Gordon CJ, Puckett E, Padnos B. Rat tail skin temperature monitored noninvasively by radiotelemetry: characterization by examination of vasomotor responses to thermomodulatory agents. J Pharmacol Toxicol Methods 47: 107–114, 2002. Gordon CJ, Watkinson WP. Behavioral and autonomic thermoregulation in the rat following chlordimeform administration. Neurotoxicol Teratol 10: 215–219, 1988. Grant RT. Vasodilation and body warming in the rat. J Physiol 167: 311–317, 1963. Han JS, Ren MF. The importance of monitoring tail-skin temperature in measuring tail-flick latency. Pain 46: 117, 1991. Hargreaves K, Dubner R, Brown F, Flores C, Joris J. A new and sensitive method for measuring thermal nociception in cutaneous hyperalgesia. Pain 32: 77– 88, 1988. Hellström B. Heat vasodilatation of the rat tail. Can J Physiol Pharmacol 53: 202–206, 1975. Hertzman AB. Some relations between skin temperature and blood flow. Am J Phys Med 32: 233–251, 1953. Hole K, Berge OG, Tjølsen A, Eide PK, Garcia-Cabrera I, Lund A, Rosland JH. The tail-flick test needs to be improved. Pain 43: 391–393, 1990. Holstein-Rathlou NH, He J, Wagner AJ, Marsh DJ. Patterns of blood pressure variability in normotensive and hypertensive rats. Am J Physiol Regul Integr Comp Physiol 269: R1230 –R1239, 1995. Jänig W. Integrative Action of the Autonomic Nervous System Neurobilogy of Homeostasis. New York: Cambridge University Press, 2006. Johnson CD, Gilbey MP. Focally recorded single sympathetic postganglionic neuronal activity supplying rat lateral tail vein. J Physiol 508: 575–585, 1998.

2197

2198

INTERACTIONS BETWEEN THERMOREGULATION AND NOCICEPTION Székely M. Skin temperature-skin blood flow: assessment of thermoregulatory changes. Acta Physiol Hung 68: 284, 1986. Tanaka M, Ootsuka Y, McKinley MJ, McAllen RM. Independent vasomotor control of rat tail and proximal hairy skin. J Physiol 582: 421– 433, 2007. The Commission for Thermal Physiology of the International Union of Physiological Sciences (IUPS Thermal Commission). Glossary of terms for thermal physiology (3rd ed.). Jpn J Physiol 51: 245–280, 2001. Thompson GE, Stevenson JA. A sex difference in the temperature response of rats to exercise. Can J Physiol Pharmacol 43: 437– 443, 1965. Thorington, RW Jr. The Biology of Rodent Tails: a Study of Form and Function (Technical report Jan 63–Feb 64). Fort Wainwright, AK: Arctic Aeromedical Laboratory, 1966, TR-65– 8, p. 1–138. Tjølsen A, Hole K. Animal models of analgesia. In: The Pharmacology of Pain. Handbook of Experimental Pharmacology, edited by Dickenson A and Besson J-M. Berlin: Springer-Verlag, 1997, vol. 130, p. 1–20. Vanhoutte G, Verhoye M, Raman E, Roberts M, Van der Linden A. In-vivo non-invasive study of the thermoregulatory function of the blood vessels in the rat tail using magnetic resonance angiography. NMR Biomed 15: 263–269, 2002. Vianna DM, Allen C, Carrive P. Cardiovascular and behavioral responses to conditioned fear after medullary raphe neuronal blockade. Neuroscience 153: 1344 –1353, 2008. Woolfe G, MacDonald AD. The evaluation of the analgesic action of pethidine hydrochloride (Demerol). J Pharmacol Exp Ther 80: 300 –307, 1944. Wu Y, Jiji LM, Lemons DE, Weinbaum S. A non-uniform three-dimensional perfusion model of rat tail heat transfer. Phys Med Biol 40: 789 – 806, 1995. Young AA, Dawson NJ. Evidence for on-off control of heat dissipation from the tail of the rat. Can J Physiol Pharmacol 60: 392–398, 1982. Zamir N, Maixner W. The relationship between cardiovascular and pain regulatory systems. Ann N Y Acad Sci 467: 371–384, 1986. Zhang YH, Yanase-Fujiwara M, Hosono T, Kanosue K. Warm and cold signals from the preoptic area: which contribute more to the control of shivering in rats? J Physiol 485: 195–202, 1995. Zimmermann M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 16: 109 –110, 1983.



Owens NC, Ootsuka Y, Kanosue K, McAllen RM. Thermoregulatory control of sympathetic fibres supplying the rat’s tail. J Physiol 543: 849 – 858, 2002. Persson PB, Baumann JE, Ehmke H, Nafz B, Wittmann U, Kirchheim HR. Phasic and 24-h blood pressure control by endothelium-derived relaxing factor in conscious dogs. Am J Physiol Heart Circ Physiol 262: H1395–H1400, 1992. Pincedé I, Pollin B, Meert T, Plaghki L, Le Bars D. Psychophysics of a nociceptive test in the mouse. PLoS One 7: e36699, 2012. Plaghki L, Decruynaere C, Van Dooren P, Le Bars D. The fine tuning of pain thresholds: a sophisticated double alarm system. PLoS One 5: e10269, 2010. Raman ER, Roberts MF, Vanhuyse VJ. Body temperature control of rat tail blood flow. Am J Physiol Regul Integr Comp Physiol 245: R426 –R432, 1983. Rand RP, Burton AC, Ing T. The tail of the rat, in temperature regulation and acclimatization. Can J Physiol Pharmacol 43: 257–267, 1965. Randich A, Maixner W. Interactions between cardiovascular and pain regulatory systems. Neurosci Biobehav Rev 8: 343–367, 1984. Rendell MS, Finnegan MF, Healy JC, Lind A, Milliken BK, Finney DE, Bonner RF. The relationship of laser-Doppler skin blood flow measurements to the cutaneous microvascular anatomy. Microvasc Res 55: 3–13, 1998. Rendell MS, McIntyre SF, Terando JV, Kelly ST, Finney DA. Skin blood flow in the Wistar-Kyoto rat and the spontaneously hypertensive rat. Comp Biochem Physiol Comp Physiol 106: 349 –354, 1993. Richardson D, Hu QF, Shepherd S. Effects of invariant sympathetic activity on cutaneous circulatory responses to heat stress. J Appl Physiol 71: 521–529, 1991. Romanovsky AA, Blatteis CM. Heat stroke: opioid-mediated mechanisms. J Appl Physiol 81: 2565–2570, 1996. Romanovsky AA, Ivanov AI, Shimansky YP. Selected contribution: ambient temperature for experiments in rats: a new method for determining the zone of thermal neutrality. J Appl Physiol 92: 2667–2679, 2002. Stauss HM. Identification of blood pressure control mechanisms by power spectral analysis. Clin Exp Pharmacol Physiol 34: 362–368, 2007.