Cutaneous Mechanical Stimulation Regulates Ovarian Blood Flow Via Activation of Spinal and Supraspinal Reflex Pathways in Anesthetized Rats

Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 1 This version is to be replaced by the final version after page-setting and proofing.

Regular Paper Cutaneous Mechanical Stimulation Regulates Ovarian Blood Flow via Activation of Spinal and Supraspinal Reflex Pathways in Anesthetized Rats

Sae UCHIDA, Fusako KAGITANI, Harumi HOTTA, Tomoko HANADA*, and Yoshihiro AIKAWA*

Department of the Autonomic Nervous System, Tokyo Metropolitan Institute of Gerontology, Itabashi-ku, Tokyo, 173-0015 Japan; and *Graduate School of Humanities and Sciences, Ochanomizu University, Bunkyo-ku, Tokyo, 112-0012, Japan

Received on Sep 30, 2005; accepted on Oct 31, 2005; released online on Nov 1, 2005; DOI: 10.2170/jjphysiol.R2133 Correspondence should be addressed to: Sae Uchida, Department of the Autonomic Nervous System, Tokyo Metropolitan Institute of Gerontology, 35-2 Sakae-cho, Itabashi-ku, Tokyo, 173-0015 Japan. Phone: +81-3-3964-3241 (Ext. 3086), Fax: +81-3-3579-4776, E-mail: [email protected]

Abstract: The reflex effects of noxious mechanical stimulation of a hindpaw or abdominal skin on ovarian blood flow, and the reflex pathways involved in those responses were examined in anesthetized rats. Blood flow in the left ovary was measured using a laser Doppler flowmeter, and the activity of the left ovarian sympathetic nerve and mean arterial pressure (MAP) of the common carotid artery were recorded. Stimulation of the left or right hindpaw for 30 s produced marked increases in ovarian sympathetic nerve activity and MAP. Ovarian blood flow slightly decreased during the stimulation and then slightly increased after the stimulation. After the left ovarian sympathetic nerves were severed, the same stimulus produced a remarkable monophasic increase in ovarian blood flow that was explained by passive vasodilation due to a marked increase in MAP. After spinal transection at the third thoracic (T3) level, the responses of MAP, ovarian sympathetic nerve activity, and ovarian blood flow to hindpaw stimulation were nearly abolished. Stimulation of the abdomen at the right or left side for 30 s produced slight increases in ovarian sympathetic nerve activity and MAP. Ovarian blood flow slightly decreased during the stimulation and then slightly increased after the stimulation. After the ovarian sympathetic nerves were severed, the response of the ovarian blood flow changed to a monophasic increase due to an increase in MAP. After spinal transection, stimulation of the left abdomen produced a moderate increase in MAP, a remarkable increase in ovarian sympathetic nerve activity and a slight decrease in ovarian blood flow during the stimulation. In contrast, stimulation of the right abdomen produced a smaller response in ovarian sympathetic nerve activity during the stimulation while it increased the MAP to a similar degree. Ovarian blood flow slightly increased after the end of stimulation, which was explained as passive vasodilation due to the increase in MAP. In conclusion, stimulation of somatic afferents affects ovarian blood flow by inducing changes in ovarian sympathetic nerve activities and blood pressure. When stimulation was applied to a hindpaw whose segment of afferent input is far from the segment of the ovarian sympathetic nerves, it took a supraspinal reflex pathway. However, when stimulation was applied to the abdomen whose spinal segment of the afferent is close to the segment of the ovarian sympathetic nerve output, there are spinal segmental reflex pathways. The present results demonstrate that spinal reflexes depend on the laterality of the stimulus, while supraspinal reflexes do not depend on the laterality of the stimulus. [The Japanese Journal of Physiology 55(5), 2005, in press]

Key words: autonomic nervous system, ovarian blood flow, ovarian sympathetic nerve, cutaneous stimulation, rat.

The ovary is innervated by autonomic nerves in addition to being under the control of hormones (see review by Burden [1]). Histological studies in rats showed that the autonomic nerves innervating the ovary are

1 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 2 This version is to be replaced by the final version after page-setting and proofing. sympathetic nerves and vagus nerves [2–5]. Recently, we demonstrated in anesthetized rats that stimulation of the sympathetic nerve that innervates the ovary causes a reduction in ovarian blood flow, while stimulation of the vagus efferent nerve that innervates the ovary had no effect on ovarian blood flow [6]. Furthermore, we demonstrated that application of a noxious mechanical stimulus to a hindpaw produced a reduction in ovarian blood flow during the stimulation by activating the ovarian sympathetic vasoconstrictor nerve [6].

Somatic afferent stimulation produces conscious sensation, emotional responses, and various autonomic responses. These somatically-induced autonomic responses including blood flow changes are produced by activation of an autonomic efferent nerve. Some of these autonomic responses induced by somatic stimulation are reflex responses called somato-autonomic reflexes. The central reflex pathways of somato-sympathetic reflexes consist of segmental spinal reflexes and generalized supraspinal reflexes (see reviews, [7–9]). Supraspinal reflexes seem to be activated particularly when a limb afferent is stimulated, and segmental reflexes seem to be activated by stimulation of spinal segmental afferents entering the spinal cord at the level of thoracic and higher lumbar segments. In rats, the sympathetic nerve innervating the ovary emerges from the spinal cord mainly at the segments of T9 and T10 [3], while afferent inputs of the hindpaw enter the spinal cord at the level of L3–L5 [10]. This anatomical evidence suggests that the sympathetically-mediated reduction in ovarian blood flow during hindpaw stimulation takes supraspinal reflex pathways instead of segmental reflex pathways.

The present study was performed to clarify whether there are spinal and/or supraspinal reflex components in the response of ovarian blood flow to cutaneous stimulation. For this purpose, we applied noxious stimulation to a hindpaw whose afferent enters the spinal cord at the level of L3–L5, and also to the abdominal skin whose afferent enters the spinal cord at around the level of T9–T12 [11]. To examine the presence of spinal reflex pathways, we prepared spinalized rats whose spinal cord was transected at the level of T3. Furthermore, we examined the involvement of somatic afferent nerves and ovarian sympathetic efferent nerves in the somatically-induced ovarian blood flow changes.

MATERIALS AND METHODS

Thirty-seven virgin female Wistar rats, 4–10 months old (body weight, 170–240 g), were used for the present experiments. The estrous cycle of each animal was determined by monitoring vaginal smears; 5 of the 37 rats were in proestrus, 18 were in estrus, 2 were in metestrus, and 12 were in diestrus on the day of the experiment. Rats were kept in a room with a 12 h:12 h light–dark schedule, with rat chow and water provided ad libitum. This study was approved by the Animal Committee of our institution. Surgical procedures. Animals were anesthetized with urethane (1.1 g/kg, I.P.). The trachea was cannulated and respiration was artificially maintained using a respirator (Model 683, Harvard, Holliston, Massachusetts, USA). The end-tidal CO2 concentration, which was monitored by a gas monitor (1H26, NEC San-ei, Tokyo, Japan), was kept at 3–4% by controlling the respiratory volume and frequency. The systemic blood pressure was continuously recorded through a cannula in a common carotid artery with a strain gauge (TP-400T, Nihon Kohden, Tokyo, Japan). A jugular vein was cannulated for infusion of necessary solutions. The animal was immobilized by administration of gallamine triethiodide (20 mg/kg, I.V., Sigma, St. Louis, MO). The core body temperature, which was monitored in the rectum, was maintained at around 37.5°C using a body temperature control system containing a thermostatically-regulated DC current heating pad and infrared lamp (ATB-1100, Nihon Kohden, Tokyo, Japan). During the experiments, urethane (10% of the dose used for initial anesthesia) was administered I.V. every 1–2 h.

Measurement of ovarian blood flow. Blood flow in the left ovary was measured in 21 rats using a laser Doppler flowmeter (ALF21D, Advance, Tokyo, Japan) (Fig. 1B). After ventral exposure of the left ovary with a wide abdominal wall opening, the ovary itself was gently placed on a small plate. The probe (outer diameter, 1.0 mm) of the flowmeter was gently placed in contact with a cover glass (about 7 × 7 mm) that had been placed on a surface of the ovary that was devoid of any visible large vessels. Care was taken to avoid compressing the ovary. The probe of the flowmeter was fixed in place using a balancing holder (ALF-B, Advance, Tokyo, Japan).

Spinal transection. Full transection of the spinal cord was performed at the third thoracic (T3) level in 10 anesthetized rats. The systolic blood pressure was kept above 70 mmHg by injection of 4% Ficoll 70 (Amersham Biosciences, Uppsala, Sweden) after spinal transection.

2 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 3 This version is to be replaced by the final version after page-setting and proofing. Cutaneous stimulation. Cutaneous nociceptive stimulation was applied by pinching an area of approximately 1 cm2 of the skin of the left or right hindpaw or the skin of the abdominal area on the left or right side for 30 s (Fig. 1A). The force of pinching was approximately 3 kg [12].

Severance of sympathetic nerves innervating the ovary. The ovarian sympathetic nerves were denervated in 7 rats. Denervation of the left ovarian sympathetic nerves was performed by cutting the suspensory ligament and accompanying blood vessels to cut the left superior ovarian nerve, and by applying a local anesthetic, procaine (1.0%), on a small piece of cotton to the left ovarian artery approximately 10 mm proximal to the ovary to block the ovarian plexus nerve ipsilateral to the ovary that was used for blood flow recording.

Recording of efferent nerve activity from the ovarian plexus nerve fibers innervating the ovary. In 16 rats, efferent nerve discharges of the left ovarian plexus nerve were recorded (Fig. 1B). The ovarian plexus nerve running along the ovarian artery was cut at about 10 mm from the ovary and covered with warm liquid paraffin. The efferent discharge activity from the proximal part of the cut end was led through a bipolar platinum-iridium wire electrode and amplified using a preamplifier (S-0476, Nihon Kohden) with a 0.01 s time constant. Nerve discharges were counted in 5-s intervals using a computer (ATAC-3700, Nihon Kohden), and the rate of nerve discharges was recorded on a polygraph (RM-6000, Nihon Kohden). Discharge activity was continuously monitored on an oscilloscope to guard against counting any artifacts of recordings that might arise.

In our previous study, we found that the response of ovarian sympathetic nerve activity to pinching of a hindpaw in central nervous system (CNS)–intact rats was not influenced by cutting the bilateral vagus nerves [6]. Therefore, in most of the experiments in the present study, the vagus parasympathetic nerves were kept intact.

Severance of somatic afferent nerves. In 5 rats, somatic afferent nerves innervating a hindpaw and abdomen were severed. The spinal nerves between the T8 and T13 levels were separated from the surrounding tissues and cut along the spine. The femoral and sciatic nerves were cut in the femoral area.

Data analysis. The ovarian blood flow, mean arterial blood pressure (MAP) and ovarian sympathetic nerve activity were calculated in 10-s intervals and expressed as percentages of the prestimulus values. Values were expressed as means ± SEM. Statistical analysis was performed by analysis of variance (ANOVA) followed by Dunnett’s multiple comparison test. Statistical significance was set at p < 0.05.

RESULTS

Effect of hindpaw pinching in CNS-intact rats Response of ovarian blood flow and MAP to hindpaw pinching. Among 11 CNS-intact animals, the ovarian blood flow and MAP under the resting condition before applying any cutaneous stimulation were 377.3 ± 29.6 mV and 105.6 ± 4.9 mmHg, respectively.

Pinching either the left or right hindpaw for 30 s always produced an increase in MAP and a decrease in ovarian blood flow during the stimulation as shown in a representative recording in Fig. 2A. Figure 3, A and D, shows the results in 17 cases. Pinching of a hindpaw reduced ovarian blood flow to 95.3 ± 1.2% of the control level during the stimulation, and the ovarian blood flow then increased to approximately 106% of the control level at 20 to 30 s after the end of stimulation. During the stimulation, the MAP increased to a maximal blood pressure of 128.1 ± 2.8% of the control value and it remained elevated for more than 90 s after the end of stimulation. These responses were elicited upon stimulation of either the left or right hindpaw. Therefore, the data obtained by stimulation of the left or right hindpaw were combined in Fig. 3, A and D.

Severance of somatic afferent nerves. When the sciatic and femoral nerves on the same side in which stimulation was unilaterally delivered were cut proximal to the stimulus, the responses of ovarian blood flow and MAP to pinching of a hindpaw were abolished (n = 5).

Severance of ovarian sympathetic nerve. After the left ovarian sympathetic nerves were severed, pinching of the left or right hindpaw was performed. Upon pinching a hindpaw, the MAP increased, similar to the response observed before the ovarian sympathetic nerves were severed , while the ovarian blood flow

3 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 4 This version is to be replaced by the final version after page-setting and proofing. showed a remarkable monophasic increase as shown in a representative case (Fig. 2B) and in the summarized results of 8 cases (Fig. 3, B and E). The MAP reached a maximal level during the 30-s pinching stimulation of 127.9 ± 4.3%. The ovarian blood flow increased during the pinching stimulation and remained significantly elevated after the end of stimulation for approximately 60 s, reaching a maximal level of 132.8 ± 4.8%.

Response of the activity of the ovarian sympathetic nerve to hindpaw pinching. Among the 12 CNS-intact rats examined, the spontaneous efferent mass activity of the ovarian sympathetic nerve before applying any cutaneous stimulation was 407.3 ± 42.1 impulses/5 s.

As shown in Fig. 2C, the level of nerve activity was elevated during the hindpaw pinching stimulation and it remained elevated after the stimulus was removed. Figure 3C summarizes the number of impulses in 17 cases. The nerve activity during the stimulation reached a maximal level of 138.5 ± 4.5%; although it started to slightly decrease during the stimulation, it remained elevated up through 90 s after the end of stimulation. Ninety seconds after the end of stimulation, the nerve activity was still elevated at 114.8 ± 3.7%. The elevated level of nerve activity gradually returned to the control level over the next 30–60 s. This response of the ovarian sympathetic nerve was elicited by stimulation of either the left or right hindpaw.

Effect of hindpaw pinching in spinalized rats Response of ovarian blood flow and MAP to hindpaw pinching. Among 4 spinalized animals, the ovarian blood flow and MAP under the resting condition were 368.8 ± 88.6 mV and 71.6 ± 2.8 mmHg, respectively.

Stimulation of either the left or right hindpaw produced a marginal increase in MAP, reaching a maximal level of approximately 102% (Figs. 4, A and B, and 5, E and F). Stimulation of either hindpaw did not significantly change the ovarian blood flow (Figs. 4, A and B, and 5, A and B).

Response of ovarian sympathetic nerve activity to hindpaw pinching. Among the 4 spinalized rats, the spontaneous efferent mass activity of the ovarian sympathetic nerve was 377.0 ± 61.9 impulses/5s.

Pinching of the left hindpaw for 30 s produced a marginal increase in MAP, as mentioned earlier (Fig. 4C). Efferent nerve activity of the ovarian sympathetic nerve slightly increased (Fig. 4C), the difference not being statistically significant (Fig. 5C). Pinching of the right hindpaw produced a marginal increase in MAP similar to left hindpaw pinching, and it did not significantly change the level of nerve activity (Figs. 4D and 5D).

Effect of abdominal pinching in CNS-intact rats Response of ovarian blood flow and MAP to abdominal pinching. Abdominal pinching for 30 s on either the left or right side produced an increase in MAP in most cases (11 out of 17 cases), although it sometimes produced a reduction in MAP (4 out of 17 cases) or no response in MAP (2 out of 17 cases). Regardless of whether the MAP increased or decreased, ovarian blood flow decreased during the stimulation as shown in Fig. 6Aab. Among the 17 cases (Fig. 7, A and D), pinching the abdominal skin reduced ovarian blood flow during the stimulation to 94.9 ± 1.0% of the control level, although the ovarian blood flow then started to recover during the stimulation and increased 20 to 50 s after the end of pinching, reaching about 108% of the control level. Overall, among the 17 cases, the MAP increased during the stimulation and remained elevated after the end of stimulation for 90 s, reaching a maximal level of 109.3 ± 1.7%. Abdominal pinching of either the left or right side elicited these responses. The ovarian blood flow during abdominal stimulation of the left or right side reached a minimum of 95.0 ± 1.0% and 94.6 ± 2.2%, respectively.

Severance of somatic afferent nerves. When the spinal nerves were cut between the T8 and T13 levels proximal to the stimulus on the same side in which stimulation was unilaterally delivered, the responses of ovarian blood flow and MAP to abdominal pinching were abolished (n = 5).

Severance of ovarian sympathetic nerve. After the left ovarian sympathetic nerves were severed, pinching of the abdomen produced an increase in MAP in most cases and a decrease in MAP in some cases during the stimulation, as observed before the ovarian sympathetic nerves were severed. Figure 6B, a and b, shows 2 typical examples during abdominal pinching. In cases where the MAP increased during the abdominal pinching, ovarian blood flow also increased in parallel with the MAP (Fig. 6Ba). In cases where the MAP decreased during the abdominal pinching, ovarian blood flow decreased in parallel with MAP (Fig. 6Bb).

4 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 5 This version is to be replaced by the final version after page-setting and proofing. Overall, among the 15 cases (Fig. 7, B and E), the MAP increased during pinching of the abdomen and it remained elevated for 90 s after the end of stimulation; the ovarian blood flow also increased during the stimulation and remained elevated after the end of stimulation for 90 s, reaching a maximal level of 116.2 ± 3.6%.

Response of ovarian sympathetic nerve activity to abdominal pinching in CNS-intact rats. In CNS-intact rats, when the MAP increased during pinching of the abdominal skin, ovarian sympathetic nerve activity also increased (Fig. 6Ca). However, when the MAP decreased during pinching of abdominal skin, the nerve activity slightly decreased during the stimulation (Fig. 6Cb). Overall, among the 17 cases, the MAP increased during the stimulation and remained elevated after the end of stimulation for 90 s. The ovarian sympathetic nerve activity started to increase during the stimulation, reaching a maximal level of 109.6 ± 2.1%, and an elevated level of nerve activity was observed until 70 s after the end of stimulation (Fig. 7C). These responses were elicited upon abdominal stimulation of either the left or right side.

Effect of abdominal pinching in spinalized rats Responses of ovarian blood flow and MAP to abdominal pinching. In the 5 spinalized rats, pinching of the left abdomen for 30 s always produced an increase in MAP and a decrease in ovarian blood flow (Fig. 8A; summarized data in Fig. 9, A and E). During the stimulation, the MAP started to increase significantly while ovarian blood flow decreased, reaching a minimal level of 94.9 ± 1.3%. After the end of stimulation, the MAP remained elevated for 60 s, although the ovarian blood flow recovered to the original level.

Stimulation of the right abdomen produced an increase in MAP as did stimulation of the left abdomen (Figs. 8B and 9F). However, stimulation of the right abdomen produced an increase in blood flow in the left ovary, reaching a maximal level of 107.8 ± 1.6% (Figs. 8B and 9B), in contrast to its response to left abdominal stimulation.

Response of ovarian sympathetic nerve activity to abdominal pinching. Pinching of the abdominal skin on the left side of spinalized rats for 30 s produced an increase in nerve activity during the stimulation, reaching a maximal level of 129.5 ± 5.6%, as well as an increase in MAP (Figs. 8C and 9C). Pinching of the right abdominal skin for 30 s produced a similar response in the MAP as that upon left abdominal stimulation, but interestingly the response of the nerve activity, reaching only 109.5 ± 2.6%, was much less than that produced by stimulation of the left abdomen (Figs. 8D and 9D).

DISCUSSION

Ovarian vasoconstriction during hindpaw pinching. In CNS-intact rats, noxious mechanical stimulation of either the left or right hindpaw produced a marked increase in ovarian sympathetic nerve activity as well as blood pressure. Ovarian blood flow slightly decreased during the stimulation and then slightly increased after the stimulation. After denervation of the ovarian sympathetic nerves, the response of ovarian blood flow changed to a remarkable monophasic increase. This indicates that the remarkable increase in ovarian blood flow in the absence of ovarian sympathetic nerves was due to a remarkable increase in blood pressure, which would cause a passive increase in blood flow. These results confirm our previous findings [6].

The present study first examined whether the reflex centers for the responses of blood flow and ovarian sympathetic nerve activity to hindpaw pinching are located in the supraspinal structure in the brain or spinal cord using spinalized rats. After spinal transection at the T3 level, stimulation of the left or right hindpaw produced a slight increase in blood pressure. However, stimulation of either hindpaw did not significantly change ovarian sympathetic nerve activity nor ovarian blood flow. These results indicate that in spinalized rats hindpaw stimulation can produce a slight increase in blood pressure but there may not exist any spinal reflex pathways from the hindpaw afferent to the spinal ovarian sympathetic nerve preganglionic neurons; accordingly, hindpaw stimulation would exert little influence on ovarian blood flow. The slight increase in blood pressure appears to be too weak to influence ovarian blood flow.

These results may indicate that in CNS-intact rats, hindpaw afferents take a supraspinal reflex pathway to connect to the ovarian sympathetic nerves (Fig. 10A). The central reflex pathway in the spinal cord between hindpaw afferents and the ovarian sympathetic nerves is very weak or none (Fig. 10B), as demonstrated in other sympathetic nerves by Araki et al. [13] and Kimura et al. [14].

5 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 6 This version is to be replaced by the final version after page-setting and proofing. Ovarian vasoconstriction during abdominal pinching. When the central nervous system was kept intact, stimulation of the abdomen on either the left or right produced an increase in ovarian sympathetic nerve activity as well as an increase in blood pressure during and after the end of stimulation. However, the ovarian blood flow decreased during the stimulation and increased after the end of stimulation. After the ovarian sympathetic nerves were severed, stimulation of the abdomen produced a similar increase in blood pressure; however, the response of ovarian blood flow changed to a monophasic increase. The initial decrease in blood flow during the stimulation that was seen in rats with intact ovarian sympathetic nerves changed to an increase response during the stimulation in rats with severed ovarian sympathetic nerves. The increase in ovarian blood flow after the end of stimulation in rats with intact ovarian sympathetic nerves was augmented after severance of the ovarian sympathetic nerves. These results suggest that after the ovarian sympathetic nerves were severed, the elevated blood pressure probably produced passive vasodilation and consequently ovarian blood flow increased. When the ovarian sympathetic nerve was intact, abdominal pinching increased the activity of this nerve, which seemed to produce vasoconstriction and consequently reduced ovarian blood flow during abdominal stimulation, in which there was a slight increase in blood pressure. However, after the end of stimulation, the blood pressure increased and a small increase in sympathetic nerve activity resulted in a marginal increase in ovarian blood flow. In some cases, abdominal pinching produced a decrease in blood pressure in CNS-intact rats, in agreement with the report of Kaufman et al. [15]. When the MAP decreased during abdominal pinching, the level of ovarian sympathetic nerve activity slightly decreased during the stimulation and ovarian blood flow decreased whether the ovarian sympathetic nerves were intact or severed. The decrease in ovarian blood flow during the depressor response to abdominal pinching is suggested to be a passive response due to the decrease in blood pressure.

After spinal transection at T3, stimulation of the left (ipsilateral) abdomen produced a moderate increase in blood pressure, a remarkable increase in ovarian sympathetic nerve activity and a slight decrease in ovarian blood flow. The moderate increase in blood pressure seems to be weaker than the remarkable increase in ovarian sympathetic nerve activity; thus, the sympathetic vasoconstrictive effect was competitively stronger than the passive vasodilative effect of the increased blood pressure on ovarian blood flow, resulting in a slight decrease in blood flow.

In the spinalized rats, however, stimulation of the right (contralateral) abdomen produced a similar increase in blood pressure as did stimulation of the left abdomen, although it induced a much smaller increase in left ovarian sympathetic nerve activity; therefore, the ovarian blood flow showed only a slight increase after the end of stimulation. In this case, the increase in blood pressure was competitively stronger than the slight increase in ovarian sympathetic nerve activity, resulting in a marginal increase in blood flow.

Considering these data, the central reflex pathways for ovarian blood vessels produced by abdominal stimulation are shown in Fig. 10, C–E. In CNS-intact rats, abdominal stimulation takes spinal and/or supraspinal reflex pathways (Fig. 10C). After spinal transection, strong spinal segmental reflex pathways from abdominal afferents to ovarian sympathetic nerves appear (Fig. 10D). The strong spinal reflex pathways that were observed in the spinalized rats, seem to be suppressed by a supraspinal structure through a descending inhibitory pathway from the brain, which is shown as a broken line in Fig. 10C, and after spinal transection, this descending inhibitory effect was eliminated and resulted in the appearance of clear responses at the spinal segmental levels (Fig. 10D). Interestingly, when an abdominal stimulation has spinal segmental reflex pathways from an afferent to the ovarian sympathetic nerves, it was noted to have a different effect on left ovarian sympathetic nerve activity depending on whether the left or right abdominal afferent was stimulated. Stimulation of the left abdomen produced a much stronger effect on left ovarian sympathetic nerve activity than stimulation of the right abdomen (Fig. 10E). Regarding the somato-sympathetic reflex at the spinal level, it was reported in other organs that stimulation of a somatic afferent ipsilateral to the side of the sympathetic nerve that was being recorded, caused much stronger activation of the reflex pathway than stimulation of the contralateral somatic afferent [14, 16].

Central reflex pathways of somato-ovarian vascular reflexes. In conclusion, stimulation of somatic afferents has some influence on ovarian blood flow that is attributed to reflex responses in ovarian sympathetic nerve activity and blood pressure. The afferent pathway consists of the cutaneous afferent nerves innervating the area that is stimulated. The central reflex pathways of the pressor responses elicited by noxious mechanical stimulation of various segmental skins have been reported [14]. The present study showed that there are two types of reflex pathways, supraspinal and propriospinal, in the somato-ovarian vascular reflexes; whether the supraspinal or propriospinal pathway is activated depends on which cutaneous segment is stimulated. The

6 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 7 This version is to be replaced by the final version after page-setting and proofing. reflex pathway from the hindpaw afferent, which enters the spinal cord at the level of L3–L5 [10], to the ovarian sympathetic nerve which emerges from the spinal cord mainly at T9–T10 [3], is a supraspinal pathway. There are at least five segments between this pair of afferent and efferent, which seem to be too far apart to take a spinal reflex pathway. When the spinal segments of somatic afferent stimulation and the ovarian sympathetic nerve are close to each other, as tested in this study, the central reflex pathway from an abdominal afferent entering the spinal cord at around T9–T12 [11] to the ovarian sympathetic nerve takes segmental spinal reflex pathways in spinalized animals; this was also noted by Sato and Schmidt [17] in a study on cats. The abolishment of the response of ovarian blood flow to hindpaw stimulation in spinalized rats was not due to spinal shock, because in the same spinalized rats abdominal stimulation produced clear spinal reflexes.

In the present study, the spinal reflexes elicited by abdominal stimulation in spinalized rats depended on the laterality of stimulation, while the supraspinal reflexes elicited by hindpaw stimulation in CNS-intact rats did not depend on the laterality of stimulation. The responses of ovarian blood flow and ovarian sympathetic nerve activity to abdominal stimulation in CNS-intact rats did not show any laterality to pinching, suggesting activation of the supraspinal reflex pathway. However, it remains to be answered if there are two reflex pathways of spinal segmental and supraspinal reflex pathways to the ovarian sympathetic nerves when abdominal skin stimulation is delivered to CNS-intact rats.

The supraspinal reflex pathway from a somatic afferent to the ovarian sympathetic nerve which influences ovarian blood flow is quite different from the reflex pathways involved in somatically-induced blood flow changes in the uterus. The reflex pathway of the somato-uterine vascular reflex was proven to take a segmental spinal reflex pathway from a somatic afferent entering the sacral spinal cord to the sacral parasympathetic uterine nerves, resulting in vasodilation of the uterus [18]. It is interesting that although the ovary and uterus are female reproductive organs that are in close proximity to each other, the somatically activated vasomotor nerves influencing ovarian blood flow and uterine blood flow are quite different; somatic stimulation activates the sympathetic vasoconstrictor nerve to the ovary [6] and the cholinergic vasodilator nerve to the uterus [19]. Somatic stimulation also activates different central reflex pathways; it activates the supraspinal reflex pathway in the case of ovarian sympathetic nerves and the segmental spinal reflex pathway in the case of uterine parasympathetic nerves.

Significance of somato-ovarian sympathetic reflexes. The ovary has two distinct functions of ovulation and production of female hormones. Zackrisson et al. [20] reported that ligation of the ovarian artery in rats resulted in a decrease in ovarian blood flow, a decrease in the rate of ovulation induced by gonadotropin injection, and a decrease in serum progesterone concentration. Therefore, the activity of the sympathetic vasoconstrictor nerve to the ovary seems to have effects on ovulation and the production of progesterone via changes in ovarian blood flow. Furthermore, the passive change in ovarian blood flow due to changes in the mean arterial blood pressure seems to have significant effects on ovulation and the production of progesterone in addition to the effect on the change in ovarian blood flow caused by changes in sympathetic nerve activity.

Histological studies demonstrated that the ovarian sympathetic nerves innervate ovarian blood vessels as well as steroidogenic interstitial gland cells in the ovary [1, 2]. Aguado and Ojeda [21] further demonstrated that severance of ovarian sympathetic nerves reduced the secretion of estrogen and progesterone at the proestrus stage in rats without changing ovarian blood flow. These results suggest that the ovarian sympathetic nerves influence ovarian blood flow and hormonal production through completely independent pathways. Therefore, the changes in ovarian sympathetic nerve activity elicited by somatic afferent stimulation probably have an effect on ovarian hormonal production in addition to ovarian blood flow.

In the present study, we demonstrated that activation of ovarian sympathetic nerve activity by noxious stimulation to the skin at various segmental skins such as the hindpaw and abdomen, changes ovarian blood flow. In these cases, the reflex pathways were proven to take supraspinal pathways and therefore, we could prove that stimulation of different segmental skin areas in anesthetized animals had similar vasoconstrictive effects on ovarian blood flow. In conscious animals, somatic stimulation leads to conscious sensation and also emotional responses that may add a further level of complexity to somato-autonomic interaction. It would be interesting to further study whether the activity of ovarian sympathetic nerves is influenced by higher CNS structures that are involved in emotional responses, and whether these descending influences from the higher nervous system to the ovarian sympathetic nerves interact with somatic afferent stimulation.

We are grateful to Prof. Akio Sato and Prof. Yuko Sato of the University of Human Arts and Sciences for their

7 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 8 This version is to be replaced by the final version after page-setting and proofing. encouragement to complete this study. This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan (to S.U.).

REFERENCES

1. Burden HW: The adrenergic innervation of mammalian ovaries. In: Serono Symposia Publications from Raven Press, vol. 18, Catecholamines as Hormone Regulators. eds. Ben-Jonathan N, Bahr JM, and Weiner RI, Raven Press, New York, pp 261–278, 1985 2. Burden HW and Lawrence IEJr.: Experimental studies on the acetylcholinesterase-positive nerves in the ovary of the rat. Anat Rec 190: 233–242, 1978 3. Gerendai I, Tóth IE, Boldogköi Z, Medveczky I, and Halász B: Neuronal labeling in the rat brain and spinal cord from the ovary using viral transneuronal tracing technique. Neuroendocrinology 68: 244–256, 1998 4. Gerendai I, Tóth IE, Boldogköi Z, Medveczky I, and Halász B: CNS structures presumably involved in vagal control of ovarian function. J Auton Nerv Syst 80: 40–45, 2000 5. Lawrence IE Jr and Burden HW: The origin of the extrinsic adrenergic innervation to the rat ovary. Anat Rec 196: 51–59, 1980 6. Uchida S, Hotta H, Kagitani F, and Aikawa Y: Ovarian blood flow is reflexively regulated by mechanical afferent stimulation of a hindlimb in nonpregnant anesthetized rats. Auton Neurosci 106: 91–97, 2003 7. Sato A, Sato Y, and Schmidt RF: The impact of somatosensory input on autonomic functions. Rev Physiol Biochem Pharmacol 130: 1–328, 1997 8. Sato A and Schmidt RF: Somatosympathetic reflexes: afferent fibers, central pathways, discharge characteristics. Physiol Rev 53: 916–947, 1973 9. Sato A and Schmidt RF: The modulation of visceral functions by somatic afferent activity. Jpn J Physiol 37: 1–17, 1987 10. Takahashi Y and Nakajima Y: Dermatomes in the rat limbs as determined by antidromic stimulation of sensory C-fibers in spinal nerves. Pain 67: 197–202, 1996 11. Sato A, Sato Y, Suzuki A, and Uchida S: Neural mechanisms of the reflex inhibition and excitation of gastric motility elicited by acupuncture-like stimulation in anesthetized rats. Neurosci Res 18: 53–62, 1993 12. Araki T, Ito K, Kurosawa M, and Sato A: Responses of adrenal sympathetic nerve activity and catecholamine secretion to cutaneous stimulation in anesthetized rats. Neuroscience 12: 289–299, 1984 13. Araki T, Hamamoto T, Kurosawa M, and Sato A: Response of adrenal efferent nerve activity to noxious stimulation of the skin. Neurosci Lett 17: 131–135, 1980 14. Kimura A, Ohsawa H, Sato A, and Sato Y: Somatocardiovascular reflexes in anesthetized rats with the central nervous system intact or acutely spinalized at the cervical level. Neurosci Res 22: 297–305, 1995 15. Kaufman A, Sato A, Sato Y, and Sugimoto H: Reflex changes in heart rate after mechanical and thermal stimulation of the skin at various segmental levels in cats. Neuroscience 2: 103–109, 1977 16. Sato A, Sato Y, and Swenson RS: Effects of morphine on somatocardiac sympathetic reflexes in spinalized cats. J Auton Nerv Syst 12: 175–184, 1985 17. Sato A and Schmidt RF: Spinal and supraspinal components of the reflex discharges into lumbar and thoracic white rami. J Physiol (Lond) 212: 839–850, 1971 18. Hotta H, Uchida S, Shimura M, and Suzuki H: Uterine contractility and blood flow are reflexively regulated by cutaneous afferent stimulation in anesthetized rats. J Auton Nerv Syst 75: 23–31, 1999 19. Sato Y, Hotta H, Nakayama H, and Suzuki H.: Sympathetic and parasympathetic regulation of the uterine blood flow and contraction in the rat. J Auton Nerv Syst 59: 151–158, 1996

8 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 9 This version is to be replaced by the final version after page-setting and proofing. 20. Zackrisson U, Mikuni M, Peterson MC, Nilsson B, Janson P-O, and Brännström M: Evidence for the involvement of blood flow-related mechanisms in the ovulatory process of the rat. Human Reprod 15: 264–272, 2000 21. Aguado LI and Ojeda SR: Ovarian adrenergic nerves play a role in maintaining preovulatory steroid secretion. Endocrinology 114: 1944–1946, 1984

Fig. 1. Schematic diagram of the experimental procedures. A: Pinching was applied to the left or right hindpaw or the abdominal area (hatched area) at the left or right side for 30 s. B: Blood flow in the left ovary was measured using a laser Doppler flowmeter. Efferent nerve discharges were recorded from the left ovarian sympathetic nerve (ovarian plexus nerve). Ovarian blood flow and ovarian sympathetic nerve activity were recorded in different animals.

Fig. 2. Typical responses of ovarian blood flow, MAP and ovarian sympathetic nerve activity to pinching of a hindpaw in CNS-intact rats. A, B: Responses of ovarian blood flow and MAP in the ovarian sympathetic nerve-intact (A) and severed (B) conditions. C: Responses of ovarian sympathetic nerve activity and MAP. The bottom bar indicates the time of stimulation.

9 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 10 This version is to be replaced by the final version after page-setting and proofing.

Fig. 3. Summary of the responses of ovarian blood flow, MAP and ovarian sympathetic nerve activity to pinching of the left or right hindpaw in CNS-intact rats. A, B, D, E: Responses of ovarian blood flow (A, B) and MAP (D, E) in ovarian sympathetic nerve-intact (A, D; n = 17 in 11 rats) and severed (B, E: n = 8 in 7 rats) conditions. C: Response of ovarian sympathetic nerve activity (n = 17 in 12 rats). Each animal was tested 1–2 times. The ovarian blood flow, MAP and ovarian sympathetic nerve activity were calculated in 10-s intervals and were expressed as percentages of the prestimulus values (ordinates). The thin dashed vertical lines and the thick horizontal bars on the abscissa indicate the time of stimulation. Each point and vertical bar represent the mean ± SEM. The onset of pinching stimulation was set as time zero (abscissa). *p < 0.05, **p < 0.01, significantly different from the prestimulus control values using one-way repeated ANOVA followed by Dunnett’s multiple comparison test. The schema on the left illustrates the experimental preparation.

Fig. 4. Typical responses of ovarian blood flow, MAP and ovarian sympathetic nerve activity to pinching of a hindpaw in spinalized rats. A, B: Responses of ovarian blood flow and MAP. C, D: Responses of ovarian sympathetic nerve activity and MAP. A, C: stimulation of left hindpaw. B, D: stimulation of right hindpaw.

Fig. 5. Summary of the responses of ovarian blood flow (A, B), MAP (E, F) and ovarian sympathetic nerve activity (C, D) to pinching of a hindpaw in spinalized rats. A, C, E: stimulation of left hindpaw. B, D, F: stimulation of right hindpaw. n = 7 in 4 rats for blood flow and MAP; n = 4 in 4 rats for ovarian sympathetic nerve activity. Other details are as in Fig. 3.

10 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 11 This version is to be replaced by the final version after page-setting and proofing.

Fig. 6. Typical responses of ovarian blood flow, MAP and ovarian sympathetic nerve activity to pinching of the abdomen in CNS-intact rats. A, B: Responses of ovarian blood flow and MAP in ovarian sympathetic nerve-intact (A) and severed (B) conditions. C: Responses of ovarian sympathetic nerve activity and MAP. a, b in A–C: Sample recordings of two different response patterns.

Fig. 7. Summary of the responses of ovarian blood flow, MAP and ovarian sympathetic nerve activity to pinching of the left or right abdomen in CNS-intact rats. A, B, D, E: Responses of ovarian blood flow (A, B) and MAP (D, E) in ovarian sympathetic nerve-intact (A, D; n = 17 in 13 rats) and severed (B, E; n = 15 in 6 rats) conditions. C: Responses of ovarian sympathetic nerve activity (n = 17 in 10 rats). Other details are as in Fig. 3.

11 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 12 This version is to be replaced by the final version after page-setting and proofing.

Fig. 8. Typical responses of ovarian blood flow, MAP and ovarian sympathetic nerve activity to pinching of the abdomen in spinalized rats. A, B: Responses of ovarian blood flow and MAP. C, D: Responses of ovarian sympathetic nerve activity and MAP. A, C: left side stimulation. B, D: right side stimulation.

Fig. 9. Summary of the responses of ovarian blood flow (A, B), MAP (E, F) and ovarian sympathetic nerve activity (C, D) to pinching of the abdomen in spinalized rats. A, C, E: left side stimulation. B, D, F: right side stimulation. n = 8 in 5 rats for blood flow; MAP, n = 4 in 4 rats for ovarian sympathetic nerve activity. Other details are as in Fig. 3.

12 Japanese Journal of Physiology Advance Publication by J-STAGE; DOI: 10.2170/jjphysiol.R2133 13 This version is to be replaced by the final version after page-setting and proofing.

Fig. 10. Schematic diagrams of the reflex pathway to the ovarian blood vessel produced by pinching of a hindpaw (A, B) or the abdomen (C, D) in CNS-intact (A, C) and spinalized (B, D) rats. A: Pinching of a hindpaw results in constriction of ovarian blood vessels through excitation of the ovarian sympathetic nerve via a supraspinal reflex pathway. B: After spinal transection, pinching of a hindpaw does not cause ovarian vasoconstriction. C: Pinching of the abdomen results in ovarian vasoconstriction through excitation of the ovarian sympathetic nerve via a spinal and/or supraspinal reflex pathway. Spinal reflex pathways are tonically inhibited by the brain through an inhibitory descending pathway (indicated by the broken line) from the brain. D: After spinal transection, inhibitory descending pathways are eliminated and pinching of the abdomen results in ovarian vasoconstriction through excitation of the ovarian sympathetic nerve via the spinal reflex pathway. +, excitatory effect; –, inhibitory effect. E: In spinalized rats, pinching of the abdomen ipsilateral (left) to the recording side of ovarian sympathetic nerve activity elicits much stronger activation of the spinal reflex than pinching of the abdomen contralateral (right) to the recording side.