- Email: [email protected]
GLUTAMATE INJECTION INTO THE CUNEIFORM NUCLEUS IN RAT, PRODUCES CORRELATED SINGLE UNIT ACTIVITIES IN THE KOLLIKER-FUSE NUCLEUS AND CARDIOVASCULAR RESPONSES
Neuroscience 223 (2012) 439–446
GLUTAMATE INJECTION INTO THE CUNEIFORM NUCLEUS IN RAT, PRODUCES CORRELATED SINGLE UNIT ACTIVITIES IN THE KOLLIKER-FUSE NUCLEUS AND CARDIOVASCULAR RESPONSES A. NASIMI, a* M. N. SHAFEI b AND H. ALAEI a
modulation (Richter and Behbehani 1991). The CnF is connected with many regions involved in cardiovascular modulation, such as the periaqueductal gray matter (PAG), the Kolliker-Fuse nucleus (KF), the locus coeruleus, the solitary tract nucleus, the dorsal motor nucleus of vagus, the lateral and paraventricular nuclei of hypothalamus (Edwards, 1975; Bernard et al., 1989; Lam et al., 1996), the raphe magnus (Zemlan and Behbehani, 1988), and the rostral ventrolateral medulla (RVLM) (Korte et al., 1992; Lam et al., 1996; Lam and Verberne, 1997; Verberne et al., 1997). KF is a part of the parabrachial/Kolliker-Fuse complex located in the rostral pons and is implicated in several functions such as regulation of cardio-respiratory systems and pain modulation (Saper and Loewy, 1980; Korte et al., 1992; Dampney and Horiuchi, 2003; Dutschmann et al., 2004; Nag and Mokha, 2004; Song et al., 2011). The KF has widespread interconnections with brain areas involved in cardiovascular, respiratory and pain regulation, such as RVLM, nucleus tractus solitarius, CnF, raphe nucleus, PAG and the intermediolateral (IML) column of the spinal cord (Korte et al., 1992; Dampney and Horiuchi, 2003; Shafei and Nasimi 2011). Electrical stimulation of the KF increased arterial pressure and produced mild tachycardia (Korte et al., 1992). Furthermore, large numbers of neurons in the KF are activated following the stimulation of cardiac sympathetic afferents (Horiuchi et al., 1999; Guo et al., 2002). Both chemical and electrical stimulations of the CnF resulted in pressor response and increased lumbar sympathetic nerve discharge (Lam and Verberne, 1997; Lam et al., 1997). But the pathway(s) mediating these effects was not defined. There are evidences that cardiovascular effects of CnF may be mediated by the RVLM (Verberne, 1995; Lam and Verberne, 1996, 1997; Verberne et al., 1997), an area playing a major role in tonic and reflex control of the cardiovascular system. However, direct projection from the CnF to the RVLM is rare (Verberne et al., 1997). Therefore it is proposed that the cardiovascular pathway(s) of the CnF to the RVLM is indirect, relayed through other nuclei especially KF. Anatomical and histochemical studies showed that the KF receives efferents from the CnF and has strong projections to the RVLM (Korte et al., 1992; Lam et al., 1996, 1997). In a recent study we showed that blockade of the KF greatly attenuated the cardiovascular responses of the CnF to glutamate stimulation (Shafei and Nasimi, 2011).
a
Department of Physiology, Isfahan University of Medical Sciences, Isfahan, Iran
b
Department of Physiology, Mashhad University of Medical Sciences, Mashhad, Iran
Abstract—The cuneiform (CnF) and Kolliker-Fuse (KF) nuclei are implicated in several functions including regulation of cardiovascular system and pain modulation. The KF also is a potential candidate for relaying the CnF cardiovascular responses to the rostral ventrolateral medulla (RVLM). In a previous study we showed that blockade of the KF strongly attenuated the short responses and moderately attenuated the long responses to glutamate microinjection into the CnF, suggesting that the cardiovascular effects of the CnF, especially the short responses, were mediated by the KF. In the present study the cellular basis of the cardiovascular responses of the CnF and possible role of the KF in relaying them to the RVLM were explored. In one group, L-glutamate was microinjected in the CnF and the cardiovascular responses were recorded. In another group the single unit responses of the KF to L-glutamate injection into the CnF were recorded. Our results showed that chemical stimulation of the CnF with glutamate produced mainly excitatory cardiovascular and single unit responses and a minority of mixed (excitatory and inhibitory) responses. In about one fourth of the cases there were no responses to stimulation. Various patterns of each group were presented and compared between cardiovascular and single unit responses. Similarities were found between cardiovascular and single unit response patterns, suggesting a significant role of KF neurons in mediating the CnF cardiovascular responses to the RVLM. Ó 2012 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: Kolliker-Fuse nucleus, cuneiform nucleus, single unit recording, glutamate, blood pressure.
INTRODUCTION The cuneiform nucleus (CnF) is a mesencephalic nucleus involved in cardiovascular responses to stress (Korte et al., 1992; Verberne, 1995; Lam et al., 1996) and pain *Corresponding author. Tel: +98-311-7922433; fax: +98-3116688597. E-mail address: [email protected] (A. Nasimi). Abbreviations: BP, blood pressure; CnF, cuneiform nucleus; HR, heart rate; KF, Kolliker-Fuse nucleus; MAP, mean arterial pressure; PAG, periaqueductal gray; PSTH, peristimulus time histogram; RVLM, rostral ventrolateral medulla; IML, Intermediolateral column.
0306-4522/12 $36.00 Ó 2012 IBRO. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neuroscience.2012.07.041 439
440
A. Nasimi et al. / Neuroscience 223 (2012) 439–446 displayed continuously on an oscilloscope then digitized and saved in multiunit mode. Then single unit firings were isolated by a program written in this lab by A. Nasimi. The program does multiple unit recordings, then segregates each single unit exactly similar to the ordinary ‘‘window discriminators’’ but more precisely. After stable firing, spontaneous activity of the cells was recorded from KF for 5–8 min. Then, L-glutamate (0.25 M, 250 nl) was microinjected into the CnF by a 1-ll Hamilton syringe. The needle was left for 30 s after drug injection. If the animal was healthy, more than one injection or recording was done in each animal, no more than two on each side and at least 300 microns apart.
In the present study the role of KF in relaying cardiovascular effects of the CnF was evaluated by single unit recording from the KF while stimulating the CnF by glutamate. Then similarities between KF single unit and CnF cardiovascular response patterns were explored. Great similarities of these responses support mediation of the KF for the CnF cardiovascular responses.
EXPERIMENTAL PROCEDURES Animals and surgery
Data analysis
Experiments were performed on 30 male Wistar rats (200–300 g) in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC). Rats were anesthetized with urethane (Sigma, 1.4 g/kg, ip) and supplementary doses (0.7 g/kg) were given if necessary. Animal’s temperature was maintained at 37 °C with a thermostatically controlled heating pad. The trachea was intubated to ease ventilation. A polyethylene catheter (PE-10) was inserted into the left femoral artery for blood pressure (BP) recording. A stainless steel guide cannula (23 gauge) was inserted stereotaxically (Stoelting, Kiel, Wisconsin, USA) 1 mm above the CnF and was fixed by dental cement. The stereotaxic coordinates of the CnF were: 7.6– 8.3 mm caudal to bregma, 1.7–2.2 mm lateral to the sagittal suture and 5.5–6.2 mm ventral from the bregma according to the atlas of Paxinos and Watson (2005). The stereotaxic coordinates of the KF were: 1.5–2.8 mm caudal to the bregma; 0.3–0.9 mm lateral to the midline; 7.0–8.5 mm ventral to the brain surface.
After recording of data, single unit spikes were isolated from the background, and a peristimulus time histogram (PSTH) was generated from the spike times. Then the cardiovascular response patterns of the CnF and the cell-firing patterns of the KF were compared.
Histology At the end of each experiment, the micropipette was moved slightly up and down to the same point to make a small mechanical lesion at the site, then the animal was sacrificed by a high dose of the anesthetic, and was then perfused transcardially with 100 ml of 0.9% saline followed by 100 ml of 10% formalin. The brain was removed and stored in 10% formalin for at least 24 h. Frozen serial transverse sections (60 microns) of forebrain were cut and stained with Cresyl Violet 1%. The injection sites were determined according to a rat brain atlas (Paxinos and Watson 2005) under the light microscope.
Experimental protocol L-glutamate (100–200 nl) was microinjected into the CnF using a single-barreled micropipette with an internal diameter of 35– 45 lm using a pressurized air pulse applicator. The volume of injection was measured by direct observation of the fluid meniscus in the micropipette by using an ocular micrometer. For the control group the same volume of the vehicle (normal saline) was microinjected into the CnF. BP and heart rate (HR) were recorded continuously using a pressure transducer connected to polygraph (HSE, March-Hugstetten, Germany) and a computer program written in this laboratory. Extracellular action potentials were also recorded in separate experiments using a glass microelectrode pulled to a fine tip diameter (1–3 lm) and filled with NaCl solution (2–4 M). Extracellular action potentials were amplified (10,000) and filtered (0.3–3 kHz) by a preamplifier (WPI, DAM 80) and
RESULTS Microinjection of the vehicle (saline, 50–100 nl, n = 21) into the CnF did not affect the mean arterial pressure (MAP: before: 93 ± 2 mmHg vs. after: 94 ± 2 mmHg), HR (before: 342 ± 8 beats/min vs. after: 340 ± 9 beats/ min) and firing rate of the KF neurons. Action potentials of 317 mostly spontaneously active neurons were recorded from the KF while stimulating the CnF by glutamate in 30 rats. Firing rate ranged from 1 to 60 spikes/s with a mean of 15.9 ± 2.2 spikes/s.
Table 1. Summary of the cardiovascular response and KF neuronal response patterns to glutamate injection into the CnF. The remaining 0.7% of the single unit responses belonged to two neurons with pure inhibitory responses not classified as a separate group Response type
Response pattern
Figure
No response
Single unit response of the KF
Cardiovascular response of the CnF
Number
Percentage
Number
Percentage
104
32.5
9
25.8
6 11 2
17.1 31.4 5.7
Excitatory
Sharp (onset) Long Onset–long excitation Onset–delay–long excitation
1 2 3 3
51 130 10 7
16 41 3.2 2.3
Mixed
Onset–long inhibition Short inhibition–long excitation Long excitation–long inhibition–long excitation
6 6 7
8
2.6
5
1.7
5 2
14.3 5.7
317
91.3
35
100
Total
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
441
the neuron recorded from the KF. The time span of the change in firing rate was comparable to the time span of cardiovascular responses for all neurons. Long excitatory responses. Some neurons (n = 130, 41%) had a long response (more than 60 s) that started and finished slowly. This was the largest group. The response length was different in various neurons, spanning from 60 s to more than 600 s (10 min). Fig. 2 shows three sets of the long responses aligned with the comparable cardiovascular responses. Microinjection of glutamate into the CnF caused a long increase in BP and HR (HR is not shown). It also caused a long increase in firing rate of the neuron in the KF. For the time span of each cardiovascular response, comparable single unit response could be found, but the very-long-excitatory responses constitute the majority of the cardiovascular responses while constitute the minority of the single unit responses. In addition, the patterns of changes are very similar between cardiovascular and firing rate responses. Onset–long excitatory responses. Some neurons (n = 10, 3.2%) had a fast sharp response (onset) followed by a long response. Fig. 3 shows a sample of this response aligned with the comparable cardiovascular responses. Microinjection of glutamate into the CnF caused a sharp increase in BP and a sharp decrease in HR followed by a slow increase in HR. This cardiovascular response is similar to the single unit response showed in Fig. 3c. Onset–delay–long excitatory responses. Some neurons (n = 7, 2.3%) had a fast sharp (onset) response followed by a delay (25–40 s) then followed by a long excitatory response (not shown). We did not find such a cardiovascular response to the CnF stimulation by glutamate. Fig. 1. A sample of sharp single unit response of the KF to stimulation of the CnF nucleus by glutamate aligned with the comparable cardiovascular responses. Arrow shows the injection time of glutamate into the CnF. (a) Heart rate change, (b) blood pressure change, and (c) single unit response showed as PSTH.
Stimulation of the CnF by glutamate caused no response in about one fourth of the cases and produced two types of cardiovascular and single unit responses, recorded from the KF, consisted of excitatory and mixed responses. These responses and their subgroups are summarized in Table 1, and are explained below. Excitatory responses Excitatory responses consisted of following patterns. Sharp excitatory (onset) response. Some neurons (n = 51, 16%) had a fast sharp response (up to 25 s) that started and finished fast. Fig. 1 shows a sample of this response aligned with the comparable cardiovascular responses. As seen, microinjection of glutamate into the CnF caused a sharp increase in BP, a decrease in HR and a sharp increase of firing rate of
Mixed responses There were some mixed (excitatory–inhibitory) cardiovascular and single unit responses (Table 1) as follows. Onset–long inhibitory responses. Some neurons (n = 8, 2.5%) had a fast sharp (onset) response followed by a long inhibition. We did not find such a cardiovascular response to the CnF stimulation by glutamate. Fig. 4 shows a sample of this response aligned with probable comparable excitatory cardiovascular responses (short inhibition–long excitation). As seen, microinjection of glutamate into the CnF caused a sharp decrease in BP and a sharp increase in HR followed by a long excitatory response. The timings of the cardiovascular response are comparable to the single unit response showed in Fig. 4c. This inhibitory response may represent the response of an inhibitory interneuron involved in the circuit generating this response. Long excitatory–long inhibitory–long excitatory responses. This was the strangest response. These
442
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
Fig. 3. A sample of onset–long single unit response of the KF to stimulation of the CnF nucleus by glutamate aligned with the comparable cardiovascular responses. There is a fast excitatory onset response followed by a slow excitatory response. Arrow shows the injection time of glutamate into the CnF. (a) Heart rate change, (b) blood pressure change, and (c) single unit response showed as PSTH. Fig. 2. Three samples of long single unit responses (PSTH) of the KF to stimulation of the CnF nucleus by glutamate aligned with the comparable cardiovascular responses. Arrow shows the injection time of glutamate into the CnF.
neurons (n = 5, 1.7%) had a long excitatory response followed by a long inhibition then followed by another very long excitatory response. Fig. 5 shows a sample of this response aligned with the comparable cardiovascular responses. Exactly similar fluctuations are seen in BP, HR and single unit responses. It is worth mentioning that there were two neurons (n = 2, 0.7%) with a long inhibitory response. We did not find a long pure inhibitory cardiovascular response to the CnF stimulation by glutamate. These inhibitory responses may represent the response of inhibitory interneurons involved in the circuit generating the long response.
Latencies Since microinjection was done by a Hamilton syringe which takes variable time, giving exact latency is not possible, however, as seen in the figures, latencies of cardiovascular and single unit responses were comparable.
Histology Fig. 6 shows the representation of the injection sites for glutamate into the CnF (a), and the recording sites from the KF (b). The data reported here are from experiments in which the injection micropipettes and recording electrodes were histologically verified to be placed in the nuclei.
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
Fig. 4. A sample of onset excitatory followed by long inhibitory single unit response from the KF to stimulation of the CnF nucleus by glutamate aligned with a probably comparable cardiovascular response. Arrow shows the injection time of glutamate into the CnF. (a) Heart rate change, (b) blood pressure change, and (c) single unit response showed as PSTH.
DISCUSSION In the present study the role of the KF in relaying cardiovascular effects of the CnF nucleus was evaluated by single unit recording from the KF neurons while stimulating the CnF nucleus by L-glutamate. Glutamate injections in the CnF in 25.8% of the cases did not produce any cardiovascular response, suggesting that not all parts of the CnF are involved in cardiovascular control. Also 32.5% of recorded units of the KF did not respond to glutamate injection in the CnF. It is known that the CnF and KF have other functions as well, like respiratory control and processing pain and stress. Therefore it is not expected that all of the KF neurons be involved in the cardiovascular responses.
443
Fig. 5. A sample of a long excitatory followed by a long inhibitory then followed by a long excitatory single unit response from the KF to the stimulation of the CnF nucleus by glutamate aligned with the comparable cardiovascular responses. Arrow shows the injection time of glutamate into the CnF. (a) Heart rate change, (b) blood pressure change, and (c) single unit response showed as PSTH.
Correlation between the cardiovascular and single unit responses Microinjection of glutamate into the CnF produced excitatory, inhibitory and mixed responses. The response patterns sum up to 5 cardiovascular and 7 single unit response patterns recorded from the KF (Table 1). It has been reported that electrical stimulation of the CnF increases arterial pressure and sympathetic vasomotor discharge (Verberne, 1995; Lam and Verberne 1997; Verberne et al., 1997). Also, previously we showed that simulation of the CnF by L-glutamate evoked a short or a long excitatory cardiovascular response (Shafei and Nasimi, 2011). In this study, comparable to our previous findings, the pressor effects sum up to 80% of the responses and 20% of the responses were mixed (Table 1). There were great similarities, both in pattern and in timing, between cardiovascular responses of the CnF and single unit responses of the KF (Figs. 1–5). Also the percentage of each response type was comparable between cardiovascular and single unit responses
444
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
Fig. 6. Schematic coronal section of rat brain adapted from an atlas (Paxinos and Watson, 2005). (a) The injection sites of glutamate (filled circles) into the CnF. (b) The recording sites from the KF.
(Table 1). Doing both types of experiments simultaneously in the same animal, will solve some other questions for example whether we see the same pattern at the same time in the two nuclei. These results are consistent with the previous anatomical and histochemical studies showing the relation between these two nuclei (Korte et al.,1992; Lam et al., 1996). For example Korte et al. (1992) demonstrated that electrical stimulation of the CnF nucleus led to sympathoexcitatory and pressor responses that were reduced by disconnection of the CnF and the KF. Our findings also confirm our previous result that the CnF nucleus sends its sympathoexcitatory responses to the RVLM mainly through the KF. Previously we showed that short cardiovascular responses of the CnF were greatly attenuated and the long responses were moderately attenuated by reversible blockade of the KF by CoCl2 (Shafei and Nasimi, 2011). In the present study, the very-long-excitatory responses constitute the majority of the cardiovascular responses while the single unit responses constitute the minority, confirming our previous finding that the KF may not be the only route for long excitatory responses of the CnF to the RVLM. Richter and Behbehani (1991) stimulated the CnF by glutamate and recorded from neurons of the raphe. They found sharp excitatory and inhibitory responses to glutamate. Since they did not find single unit responses similar to the cardiovascular responses of the CnF, the
raphe magnus may not mediate the cardiovascular responses of the CnF. Overall, the KF is a major relaying site of cardiovascular responses of the CnF to the RVLM. High similarities between cardiovascular and single unit responses also demonstrate that the KF signals are not (or little) modulated in the RVLM. The KF signals originating from the CnF, directly stimulate RVLM premotor sympathoexcitatory neurons (Verberne et al., 1997). Possible advantages of the CnF responses The CnF is a part of defense (Mitchell et al., 1988) and pain modulation systems and its projections into PAG and raphe magnus nucleus, important components of the descending pain modulatory system, are well known (Zemlan and Behbehani, 1988; Bernard et al., 1989; Verberne et al., 1997; Haghparast et al., 2007). It has also been shown that pain evokes cardiovascular responses such as increases in MAP, HR, and myocardial contractility (Siddall et al., 1994; Karlsson et al., 2006). The CnF by sending axons into the raphe magnus and PAG decreases the pain sensation (Zemlan and Behbehani, 1988) and simultaneously by projecting into the RVLM premotor sympathoexcitatory neurons and the intermediolateral cell column (Korte et al., 1992) increases BP to help the animal to defend against the pain-generating factors. However in 25.8% of the cases
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
it produces no responses and in 5.7% of the times the CnF produces long oscillation of BP (Table 1, Fig. 5). Single unit response patterns We found that single unit response patterns are distributed throughout the CnF as micro-clusters. In other words the neurons near to each other usually show similar pattern. Sometimes a shift from one cluster to another was seen. Therefore based on the stimulated cluster, a special pattern of cardiovascular response must be seen. Since it was not possible to histologically designate a part of the nucleus to a specific response pattern, it seems that the functional clusters are not grouped in specific parts of the nucleus. We found that the long responses shorter than 150 s comprise 5.7% of the cardiovascular responses while they comprise 30% of the single unit responses, suggesting that there is a tendency to sum the shorter responses to very long responses before exiting the CnF. In support of that, in some single unit recordings we found a regular pattern of increase in the rate and duration in the KF that might be due to summations. Our results showed that there are inhibitory single unit responses (Table 1, Figs. 4 and 5) to glutamate microinjection as well as excitatory responses. These responses could have arisen from GABAergic or other inhibitory interneurons stimulated by glutamate and might help to modulate or switch off the excitatory responses. Some single unit response patterns are resulted from convergence of both excitatory and inhibitory signals on the same neuron. For example in Fig. 4, the neurons must have received a short inhibitory response followed by an incoming excitatory signal. Fig. 4c shows the response of a possible inhibitory neuron that might have produced the onset inhibition of the cardiovascular response. In Fig. 5, successive excitatory, inhibitory, excitatory signals converge on the same neuron resulting in disadvantageous oscillation of BP. Finally based on the present data one could not be sure whether the suggested circuits are located in the CnF or in the KF. Similar recordings should be done in the CnF to find the answer. In summary, our results showed that chemical stimulation of the CnF with glutamate produced 5 cardiovascular and 6 single unit response patterns recorded from the KF (Table 1). Similarities between cardiovascular and single unit response patterns, suggest a significant role of KF neurons in mediating the CnF cardiovascular responses to the RVLM. Acknowledgements—This study was supported by a grant from Vice-Chancellery of Research of the Isfahan University of Medical Sciences.
REFERENCES Bernard JF, Peschanski M, Besson JM (1989) Afferents and efferents of the rat cuneiformis nucleus: an anatomical study with reference to pain transmission. Brain Res 490(1):181–185.
445
Dampney RA, Horiuchi J (2003) Functional organisation of central cardiovascular pathways: studies using c-fos gene expression. Prog Neurobiol 71(5):359–384. Dutschmann M, Morschel M, Kron M, Herbert H (2004) Development of adaptive behaviour of the respiratory network: implications for the pontine Ko¨lliker-Fuse nucleus. Respir Physiol Neurobiol 143(2–3):155–165. Edwards SB (1975) Autoradiographic studies of the projections of the midbrain reticular formation, descending projections of nucleus cuneiformis. J Comp Neurol 161(3):341–358. Guo ZL, Li P, Longhurst J (2002) Central pathways in the pons and midbrain involved in cardiac sympathoexcitatory reflexes in cats. Neuroscience 113(2):435–447. Haghparast A, Soltani-Hekmat A, Khani A, Komaki A (2007) Role of glutamatergic receptors located in the nucleus raphe magnus on antinociceptive effect of morphine microinjected into the nucleus cuneiformis of rat. Neurosci Lett 427(1):44–49. Horiuchi J, Potts P, Polson J, Dampney R (1999) Distribution of neurons projecting to the rostral ventrolateral medullary pressor region that are activated by sustained hypotension. Neuroscience 89(4):1319–1329. Karlsson GA, Preuss CV, Chaitoff KA, Maher TJ, Ally A (2006) Medullary monoamines and NMDA-receptor regulation of cardiovascular responses during peripheral nociceptive stimuli. Neurosci Res 55(3):316–326. Korte SM, Jaarsma D, Luiten PG, Bohus B (1992) Mesencephalic cuneiform nucleus and its ascending and descending projections serve stress-related cardiovascular responses in the rat. J Auton Nerv Syst 41(1–2):157–176. Lam W, Gundlach AL, Verberne AJ (1996) Increased nerve growth factor inducible-A gene and c-fos messenger RNA levels in the rat midbrain and hindbrain associated with the cardiovascular response to electrical stimulation of the mesencephalic cuneiform nucleus. Neuroscience 71(1):193–211. Lam W, Gundlach AL, Verberne AJ (1997) Neuronal activation in the forebrain following electrical stimulation of the cuneiform nucleus in the rat: hypothalamic expression of c-fos and NGFI-A messenger RNA. Neuroscience 78(4):1069–1085. Lam W, Verberne AJ (1997) Cuneiform nucleus stimulation-induced sympathoexcitation: role of adrenoceptors, excitatory amino acid and serotonin receptors in rat spinal cord. Brain Res 757(2): 191–201. Mitchell IJ, Dean P, Redgrave P (1988) The projection from superior colliculus to cuneiform area in the rat. II. Defencelike responses to stimulation with glutamate in cuneiform nucleus and surrounding structures. Exp Brain Res 72(3): 626–639. Nag S, Mokha SS (2004) Estrogen attenuates antinociception produced by stimulation of Kolliker-Fuse nucleus in the rat. Eur J Neurosci 20(11):3203–3207. Paxinos G, Watson C (2005) The rat brain in stereotaxic coordinates. London, UK: Elsevier. Richter RC, Behbehani MM (1991) Evidence for glutamic acid as a possible neurotransmitter between the mesencephalic nucleus cuneiformis and the medullary nucleus raphe magnus in the lightly anesthetized rat. Brain Res 544:279–286. Saper CB, Loewy AD (1980) Efferent connections of the parabrachial nucleus in the rat. Brain Res 197(2):291–317. Shafei MN, Nasimi A (2011) Effect of glutamate stimulation of the cuneiform nucleus on cardiovascular regulation in anesthetized rats: Role of the pontine Kolliker-Fuse nucleus. Brain Res 1385: 135–143. Siddall PJ, Polson JW, Dampney RA (1994) Descending antinociceptive pathway from the rostral ventrolateral medulla: a correlative anatomical and physiological study. Brain Res 645(1– 2):61–68. Song G, Xu H, Wang H, Macdonald SM, Poon CS (2011) Hypoxiaexcited neurons in NTS send axonal projections to Kolliker-Fuse/ parabrachial complex in dorsolateral pons. Neuroscience 175:145–153.
446
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
Verberne AJ, Lam W, Owens NC, Sartor D (1997) Supramedullary modulation of sympathetic vasomotor function. Clin Exp Pharmacol Physiol 24(9–10):748–754. Verberne AJ (1995) Cuneiform nucleus stimulation produces activation of medullary sympathoexcitatory neurons in rats. Am J Physiol 268(3 Pt 2):R752–R758.
Zemlan FP, Behbehani MM (1988) Nucleus cuneiformis and pain modulation: anatomy and behavioral pharmacology. Brain Res 453(1–2):89–102.
(Accepted 24 July 2012) (Available online 31 July 2012)
GLUTAMATE INJECTION INTO THE CUNEIFORM NUCLEUS IN RAT, PRODUCES CORRELATED SINGLE UNIT ACTIVITIES IN THE KOLLIKER-FUSE NUCLEUS AND CARDIOVASCULAR RESPONSES A. NASIMI, a* M. N. SHAFEI b AND H. ALAEI a
modulation (Richter and Behbehani 1991). The CnF is connected with many regions involved in cardiovascular modulation, such as the periaqueductal gray matter (PAG), the Kolliker-Fuse nucleus (KF), the locus coeruleus, the solitary tract nucleus, the dorsal motor nucleus of vagus, the lateral and paraventricular nuclei of hypothalamus (Edwards, 1975; Bernard et al., 1989; Lam et al., 1996), the raphe magnus (Zemlan and Behbehani, 1988), and the rostral ventrolateral medulla (RVLM) (Korte et al., 1992; Lam et al., 1996; Lam and Verberne, 1997; Verberne et al., 1997). KF is a part of the parabrachial/Kolliker-Fuse complex located in the rostral pons and is implicated in several functions such as regulation of cardio-respiratory systems and pain modulation (Saper and Loewy, 1980; Korte et al., 1992; Dampney and Horiuchi, 2003; Dutschmann et al., 2004; Nag and Mokha, 2004; Song et al., 2011). The KF has widespread interconnections with brain areas involved in cardiovascular, respiratory and pain regulation, such as RVLM, nucleus tractus solitarius, CnF, raphe nucleus, PAG and the intermediolateral (IML) column of the spinal cord (Korte et al., 1992; Dampney and Horiuchi, 2003; Shafei and Nasimi 2011). Electrical stimulation of the KF increased arterial pressure and produced mild tachycardia (Korte et al., 1992). Furthermore, large numbers of neurons in the KF are activated following the stimulation of cardiac sympathetic afferents (Horiuchi et al., 1999; Guo et al., 2002). Both chemical and electrical stimulations of the CnF resulted in pressor response and increased lumbar sympathetic nerve discharge (Lam and Verberne, 1997; Lam et al., 1997). But the pathway(s) mediating these effects was not defined. There are evidences that cardiovascular effects of CnF may be mediated by the RVLM (Verberne, 1995; Lam and Verberne, 1996, 1997; Verberne et al., 1997), an area playing a major role in tonic and reflex control of the cardiovascular system. However, direct projection from the CnF to the RVLM is rare (Verberne et al., 1997). Therefore it is proposed that the cardiovascular pathway(s) of the CnF to the RVLM is indirect, relayed through other nuclei especially KF. Anatomical and histochemical studies showed that the KF receives efferents from the CnF and has strong projections to the RVLM (Korte et al., 1992; Lam et al., 1996, 1997). In a recent study we showed that blockade of the KF greatly attenuated the cardiovascular responses of the CnF to glutamate stimulation (Shafei and Nasimi, 2011).
a
Department of Physiology, Isfahan University of Medical Sciences, Isfahan, Iran
b
Department of Physiology, Mashhad University of Medical Sciences, Mashhad, Iran
Abstract—The cuneiform (CnF) and Kolliker-Fuse (KF) nuclei are implicated in several functions including regulation of cardiovascular system and pain modulation. The KF also is a potential candidate for relaying the CnF cardiovascular responses to the rostral ventrolateral medulla (RVLM). In a previous study we showed that blockade of the KF strongly attenuated the short responses and moderately attenuated the long responses to glutamate microinjection into the CnF, suggesting that the cardiovascular effects of the CnF, especially the short responses, were mediated by the KF. In the present study the cellular basis of the cardiovascular responses of the CnF and possible role of the KF in relaying them to the RVLM were explored. In one group, L-glutamate was microinjected in the CnF and the cardiovascular responses were recorded. In another group the single unit responses of the KF to L-glutamate injection into the CnF were recorded. Our results showed that chemical stimulation of the CnF with glutamate produced mainly excitatory cardiovascular and single unit responses and a minority of mixed (excitatory and inhibitory) responses. In about one fourth of the cases there were no responses to stimulation. Various patterns of each group were presented and compared between cardiovascular and single unit responses. Similarities were found between cardiovascular and single unit response patterns, suggesting a significant role of KF neurons in mediating the CnF cardiovascular responses to the RVLM. Ó 2012 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: Kolliker-Fuse nucleus, cuneiform nucleus, single unit recording, glutamate, blood pressure.
INTRODUCTION The cuneiform nucleus (CnF) is a mesencephalic nucleus involved in cardiovascular responses to stress (Korte et al., 1992; Verberne, 1995; Lam et al., 1996) and pain *Corresponding author. Tel: +98-311-7922433; fax: +98-3116688597. E-mail address: [email protected] (A. Nasimi). Abbreviations: BP, blood pressure; CnF, cuneiform nucleus; HR, heart rate; KF, Kolliker-Fuse nucleus; MAP, mean arterial pressure; PAG, periaqueductal gray; PSTH, peristimulus time histogram; RVLM, rostral ventrolateral medulla; IML, Intermediolateral column.
0306-4522/12 $36.00 Ó 2012 IBRO. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neuroscience.2012.07.041 439
440
A. Nasimi et al. / Neuroscience 223 (2012) 439–446 displayed continuously on an oscilloscope then digitized and saved in multiunit mode. Then single unit firings were isolated by a program written in this lab by A. Nasimi. The program does multiple unit recordings, then segregates each single unit exactly similar to the ordinary ‘‘window discriminators’’ but more precisely. After stable firing, spontaneous activity of the cells was recorded from KF for 5–8 min. Then, L-glutamate (0.25 M, 250 nl) was microinjected into the CnF by a 1-ll Hamilton syringe. The needle was left for 30 s after drug injection. If the animal was healthy, more than one injection or recording was done in each animal, no more than two on each side and at least 300 microns apart.
In the present study the role of KF in relaying cardiovascular effects of the CnF was evaluated by single unit recording from the KF while stimulating the CnF by glutamate. Then similarities between KF single unit and CnF cardiovascular response patterns were explored. Great similarities of these responses support mediation of the KF for the CnF cardiovascular responses.
EXPERIMENTAL PROCEDURES Animals and surgery
Data analysis
Experiments were performed on 30 male Wistar rats (200–300 g) in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC). Rats were anesthetized with urethane (Sigma, 1.4 g/kg, ip) and supplementary doses (0.7 g/kg) were given if necessary. Animal’s temperature was maintained at 37 °C with a thermostatically controlled heating pad. The trachea was intubated to ease ventilation. A polyethylene catheter (PE-10) was inserted into the left femoral artery for blood pressure (BP) recording. A stainless steel guide cannula (23 gauge) was inserted stereotaxically (Stoelting, Kiel, Wisconsin, USA) 1 mm above the CnF and was fixed by dental cement. The stereotaxic coordinates of the CnF were: 7.6– 8.3 mm caudal to bregma, 1.7–2.2 mm lateral to the sagittal suture and 5.5–6.2 mm ventral from the bregma according to the atlas of Paxinos and Watson (2005). The stereotaxic coordinates of the KF were: 1.5–2.8 mm caudal to the bregma; 0.3–0.9 mm lateral to the midline; 7.0–8.5 mm ventral to the brain surface.
After recording of data, single unit spikes were isolated from the background, and a peristimulus time histogram (PSTH) was generated from the spike times. Then the cardiovascular response patterns of the CnF and the cell-firing patterns of the KF were compared.
Histology At the end of each experiment, the micropipette was moved slightly up and down to the same point to make a small mechanical lesion at the site, then the animal was sacrificed by a high dose of the anesthetic, and was then perfused transcardially with 100 ml of 0.9% saline followed by 100 ml of 10% formalin. The brain was removed and stored in 10% formalin for at least 24 h. Frozen serial transverse sections (60 microns) of forebrain were cut and stained with Cresyl Violet 1%. The injection sites were determined according to a rat brain atlas (Paxinos and Watson 2005) under the light microscope.
Experimental protocol L-glutamate (100–200 nl) was microinjected into the CnF using a single-barreled micropipette with an internal diameter of 35– 45 lm using a pressurized air pulse applicator. The volume of injection was measured by direct observation of the fluid meniscus in the micropipette by using an ocular micrometer. For the control group the same volume of the vehicle (normal saline) was microinjected into the CnF. BP and heart rate (HR) were recorded continuously using a pressure transducer connected to polygraph (HSE, March-Hugstetten, Germany) and a computer program written in this laboratory. Extracellular action potentials were also recorded in separate experiments using a glass microelectrode pulled to a fine tip diameter (1–3 lm) and filled with NaCl solution (2–4 M). Extracellular action potentials were amplified (10,000) and filtered (0.3–3 kHz) by a preamplifier (WPI, DAM 80) and
RESULTS Microinjection of the vehicle (saline, 50–100 nl, n = 21) into the CnF did not affect the mean arterial pressure (MAP: before: 93 ± 2 mmHg vs. after: 94 ± 2 mmHg), HR (before: 342 ± 8 beats/min vs. after: 340 ± 9 beats/ min) and firing rate of the KF neurons. Action potentials of 317 mostly spontaneously active neurons were recorded from the KF while stimulating the CnF by glutamate in 30 rats. Firing rate ranged from 1 to 60 spikes/s with a mean of 15.9 ± 2.2 spikes/s.
Table 1. Summary of the cardiovascular response and KF neuronal response patterns to glutamate injection into the CnF. The remaining 0.7% of the single unit responses belonged to two neurons with pure inhibitory responses not classified as a separate group Response type
Response pattern
Figure
No response
Single unit response of the KF
Cardiovascular response of the CnF
Number
Percentage
Number
Percentage
104
32.5
9
25.8
6 11 2
17.1 31.4 5.7
Excitatory
Sharp (onset) Long Onset–long excitation Onset–delay–long excitation
1 2 3 3
51 130 10 7
16 41 3.2 2.3
Mixed
Onset–long inhibition Short inhibition–long excitation Long excitation–long inhibition–long excitation
6 6 7
8
2.6
5
1.7
5 2
14.3 5.7
317
91.3
35
100
Total
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
441
the neuron recorded from the KF. The time span of the change in firing rate was comparable to the time span of cardiovascular responses for all neurons. Long excitatory responses. Some neurons (n = 130, 41%) had a long response (more than 60 s) that started and finished slowly. This was the largest group. The response length was different in various neurons, spanning from 60 s to more than 600 s (10 min). Fig. 2 shows three sets of the long responses aligned with the comparable cardiovascular responses. Microinjection of glutamate into the CnF caused a long increase in BP and HR (HR is not shown). It also caused a long increase in firing rate of the neuron in the KF. For the time span of each cardiovascular response, comparable single unit response could be found, but the very-long-excitatory responses constitute the majority of the cardiovascular responses while constitute the minority of the single unit responses. In addition, the patterns of changes are very similar between cardiovascular and firing rate responses. Onset–long excitatory responses. Some neurons (n = 10, 3.2%) had a fast sharp response (onset) followed by a long response. Fig. 3 shows a sample of this response aligned with the comparable cardiovascular responses. Microinjection of glutamate into the CnF caused a sharp increase in BP and a sharp decrease in HR followed by a slow increase in HR. This cardiovascular response is similar to the single unit response showed in Fig. 3c. Onset–delay–long excitatory responses. Some neurons (n = 7, 2.3%) had a fast sharp (onset) response followed by a delay (25–40 s) then followed by a long excitatory response (not shown). We did not find such a cardiovascular response to the CnF stimulation by glutamate. Fig. 1. A sample of sharp single unit response of the KF to stimulation of the CnF nucleus by glutamate aligned with the comparable cardiovascular responses. Arrow shows the injection time of glutamate into the CnF. (a) Heart rate change, (b) blood pressure change, and (c) single unit response showed as PSTH.
Stimulation of the CnF by glutamate caused no response in about one fourth of the cases and produced two types of cardiovascular and single unit responses, recorded from the KF, consisted of excitatory and mixed responses. These responses and their subgroups are summarized in Table 1, and are explained below. Excitatory responses Excitatory responses consisted of following patterns. Sharp excitatory (onset) response. Some neurons (n = 51, 16%) had a fast sharp response (up to 25 s) that started and finished fast. Fig. 1 shows a sample of this response aligned with the comparable cardiovascular responses. As seen, microinjection of glutamate into the CnF caused a sharp increase in BP, a decrease in HR and a sharp increase of firing rate of
Mixed responses There were some mixed (excitatory–inhibitory) cardiovascular and single unit responses (Table 1) as follows. Onset–long inhibitory responses. Some neurons (n = 8, 2.5%) had a fast sharp (onset) response followed by a long inhibition. We did not find such a cardiovascular response to the CnF stimulation by glutamate. Fig. 4 shows a sample of this response aligned with probable comparable excitatory cardiovascular responses (short inhibition–long excitation). As seen, microinjection of glutamate into the CnF caused a sharp decrease in BP and a sharp increase in HR followed by a long excitatory response. The timings of the cardiovascular response are comparable to the single unit response showed in Fig. 4c. This inhibitory response may represent the response of an inhibitory interneuron involved in the circuit generating this response. Long excitatory–long inhibitory–long excitatory responses. This was the strangest response. These
442
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
Fig. 3. A sample of onset–long single unit response of the KF to stimulation of the CnF nucleus by glutamate aligned with the comparable cardiovascular responses. There is a fast excitatory onset response followed by a slow excitatory response. Arrow shows the injection time of glutamate into the CnF. (a) Heart rate change, (b) blood pressure change, and (c) single unit response showed as PSTH. Fig. 2. Three samples of long single unit responses (PSTH) of the KF to stimulation of the CnF nucleus by glutamate aligned with the comparable cardiovascular responses. Arrow shows the injection time of glutamate into the CnF.
neurons (n = 5, 1.7%) had a long excitatory response followed by a long inhibition then followed by another very long excitatory response. Fig. 5 shows a sample of this response aligned with the comparable cardiovascular responses. Exactly similar fluctuations are seen in BP, HR and single unit responses. It is worth mentioning that there were two neurons (n = 2, 0.7%) with a long inhibitory response. We did not find a long pure inhibitory cardiovascular response to the CnF stimulation by glutamate. These inhibitory responses may represent the response of inhibitory interneurons involved in the circuit generating the long response.
Latencies Since microinjection was done by a Hamilton syringe which takes variable time, giving exact latency is not possible, however, as seen in the figures, latencies of cardiovascular and single unit responses were comparable.
Histology Fig. 6 shows the representation of the injection sites for glutamate into the CnF (a), and the recording sites from the KF (b). The data reported here are from experiments in which the injection micropipettes and recording electrodes were histologically verified to be placed in the nuclei.
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
Fig. 4. A sample of onset excitatory followed by long inhibitory single unit response from the KF to stimulation of the CnF nucleus by glutamate aligned with a probably comparable cardiovascular response. Arrow shows the injection time of glutamate into the CnF. (a) Heart rate change, (b) blood pressure change, and (c) single unit response showed as PSTH.
DISCUSSION In the present study the role of the KF in relaying cardiovascular effects of the CnF nucleus was evaluated by single unit recording from the KF neurons while stimulating the CnF nucleus by L-glutamate. Glutamate injections in the CnF in 25.8% of the cases did not produce any cardiovascular response, suggesting that not all parts of the CnF are involved in cardiovascular control. Also 32.5% of recorded units of the KF did not respond to glutamate injection in the CnF. It is known that the CnF and KF have other functions as well, like respiratory control and processing pain and stress. Therefore it is not expected that all of the KF neurons be involved in the cardiovascular responses.
443
Fig. 5. A sample of a long excitatory followed by a long inhibitory then followed by a long excitatory single unit response from the KF to the stimulation of the CnF nucleus by glutamate aligned with the comparable cardiovascular responses. Arrow shows the injection time of glutamate into the CnF. (a) Heart rate change, (b) blood pressure change, and (c) single unit response showed as PSTH.
Correlation between the cardiovascular and single unit responses Microinjection of glutamate into the CnF produced excitatory, inhibitory and mixed responses. The response patterns sum up to 5 cardiovascular and 7 single unit response patterns recorded from the KF (Table 1). It has been reported that electrical stimulation of the CnF increases arterial pressure and sympathetic vasomotor discharge (Verberne, 1995; Lam and Verberne 1997; Verberne et al., 1997). Also, previously we showed that simulation of the CnF by L-glutamate evoked a short or a long excitatory cardiovascular response (Shafei and Nasimi, 2011). In this study, comparable to our previous findings, the pressor effects sum up to 80% of the responses and 20% of the responses were mixed (Table 1). There were great similarities, both in pattern and in timing, between cardiovascular responses of the CnF and single unit responses of the KF (Figs. 1–5). Also the percentage of each response type was comparable between cardiovascular and single unit responses
444
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
Fig. 6. Schematic coronal section of rat brain adapted from an atlas (Paxinos and Watson, 2005). (a) The injection sites of glutamate (filled circles) into the CnF. (b) The recording sites from the KF.
(Table 1). Doing both types of experiments simultaneously in the same animal, will solve some other questions for example whether we see the same pattern at the same time in the two nuclei. These results are consistent with the previous anatomical and histochemical studies showing the relation between these two nuclei (Korte et al.,1992; Lam et al., 1996). For example Korte et al. (1992) demonstrated that electrical stimulation of the CnF nucleus led to sympathoexcitatory and pressor responses that were reduced by disconnection of the CnF and the KF. Our findings also confirm our previous result that the CnF nucleus sends its sympathoexcitatory responses to the RVLM mainly through the KF. Previously we showed that short cardiovascular responses of the CnF were greatly attenuated and the long responses were moderately attenuated by reversible blockade of the KF by CoCl2 (Shafei and Nasimi, 2011). In the present study, the very-long-excitatory responses constitute the majority of the cardiovascular responses while the single unit responses constitute the minority, confirming our previous finding that the KF may not be the only route for long excitatory responses of the CnF to the RVLM. Richter and Behbehani (1991) stimulated the CnF by glutamate and recorded from neurons of the raphe. They found sharp excitatory and inhibitory responses to glutamate. Since they did not find single unit responses similar to the cardiovascular responses of the CnF, the
raphe magnus may not mediate the cardiovascular responses of the CnF. Overall, the KF is a major relaying site of cardiovascular responses of the CnF to the RVLM. High similarities between cardiovascular and single unit responses also demonstrate that the KF signals are not (or little) modulated in the RVLM. The KF signals originating from the CnF, directly stimulate RVLM premotor sympathoexcitatory neurons (Verberne et al., 1997). Possible advantages of the CnF responses The CnF is a part of defense (Mitchell et al., 1988) and pain modulation systems and its projections into PAG and raphe magnus nucleus, important components of the descending pain modulatory system, are well known (Zemlan and Behbehani, 1988; Bernard et al., 1989; Verberne et al., 1997; Haghparast et al., 2007). It has also been shown that pain evokes cardiovascular responses such as increases in MAP, HR, and myocardial contractility (Siddall et al., 1994; Karlsson et al., 2006). The CnF by sending axons into the raphe magnus and PAG decreases the pain sensation (Zemlan and Behbehani, 1988) and simultaneously by projecting into the RVLM premotor sympathoexcitatory neurons and the intermediolateral cell column (Korte et al., 1992) increases BP to help the animal to defend against the pain-generating factors. However in 25.8% of the cases
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
it produces no responses and in 5.7% of the times the CnF produces long oscillation of BP (Table 1, Fig. 5). Single unit response patterns We found that single unit response patterns are distributed throughout the CnF as micro-clusters. In other words the neurons near to each other usually show similar pattern. Sometimes a shift from one cluster to another was seen. Therefore based on the stimulated cluster, a special pattern of cardiovascular response must be seen. Since it was not possible to histologically designate a part of the nucleus to a specific response pattern, it seems that the functional clusters are not grouped in specific parts of the nucleus. We found that the long responses shorter than 150 s comprise 5.7% of the cardiovascular responses while they comprise 30% of the single unit responses, suggesting that there is a tendency to sum the shorter responses to very long responses before exiting the CnF. In support of that, in some single unit recordings we found a regular pattern of increase in the rate and duration in the KF that might be due to summations. Our results showed that there are inhibitory single unit responses (Table 1, Figs. 4 and 5) to glutamate microinjection as well as excitatory responses. These responses could have arisen from GABAergic or other inhibitory interneurons stimulated by glutamate and might help to modulate or switch off the excitatory responses. Some single unit response patterns are resulted from convergence of both excitatory and inhibitory signals on the same neuron. For example in Fig. 4, the neurons must have received a short inhibitory response followed by an incoming excitatory signal. Fig. 4c shows the response of a possible inhibitory neuron that might have produced the onset inhibition of the cardiovascular response. In Fig. 5, successive excitatory, inhibitory, excitatory signals converge on the same neuron resulting in disadvantageous oscillation of BP. Finally based on the present data one could not be sure whether the suggested circuits are located in the CnF or in the KF. Similar recordings should be done in the CnF to find the answer. In summary, our results showed that chemical stimulation of the CnF with glutamate produced 5 cardiovascular and 6 single unit response patterns recorded from the KF (Table 1). Similarities between cardiovascular and single unit response patterns, suggest a significant role of KF neurons in mediating the CnF cardiovascular responses to the RVLM. Acknowledgements—This study was supported by a grant from Vice-Chancellery of Research of the Isfahan University of Medical Sciences.
REFERENCES Bernard JF, Peschanski M, Besson JM (1989) Afferents and efferents of the rat cuneiformis nucleus: an anatomical study with reference to pain transmission. Brain Res 490(1):181–185.
445
Dampney RA, Horiuchi J (2003) Functional organisation of central cardiovascular pathways: studies using c-fos gene expression. Prog Neurobiol 71(5):359–384. Dutschmann M, Morschel M, Kron M, Herbert H (2004) Development of adaptive behaviour of the respiratory network: implications for the pontine Ko¨lliker-Fuse nucleus. Respir Physiol Neurobiol 143(2–3):155–165. Edwards SB (1975) Autoradiographic studies of the projections of the midbrain reticular formation, descending projections of nucleus cuneiformis. J Comp Neurol 161(3):341–358. Guo ZL, Li P, Longhurst J (2002) Central pathways in the pons and midbrain involved in cardiac sympathoexcitatory reflexes in cats. Neuroscience 113(2):435–447. Haghparast A, Soltani-Hekmat A, Khani A, Komaki A (2007) Role of glutamatergic receptors located in the nucleus raphe magnus on antinociceptive effect of morphine microinjected into the nucleus cuneiformis of rat. Neurosci Lett 427(1):44–49. Horiuchi J, Potts P, Polson J, Dampney R (1999) Distribution of neurons projecting to the rostral ventrolateral medullary pressor region that are activated by sustained hypotension. Neuroscience 89(4):1319–1329. Karlsson GA, Preuss CV, Chaitoff KA, Maher TJ, Ally A (2006) Medullary monoamines and NMDA-receptor regulation of cardiovascular responses during peripheral nociceptive stimuli. Neurosci Res 55(3):316–326. Korte SM, Jaarsma D, Luiten PG, Bohus B (1992) Mesencephalic cuneiform nucleus and its ascending and descending projections serve stress-related cardiovascular responses in the rat. J Auton Nerv Syst 41(1–2):157–176. Lam W, Gundlach AL, Verberne AJ (1996) Increased nerve growth factor inducible-A gene and c-fos messenger RNA levels in the rat midbrain and hindbrain associated with the cardiovascular response to electrical stimulation of the mesencephalic cuneiform nucleus. Neuroscience 71(1):193–211. Lam W, Gundlach AL, Verberne AJ (1997) Neuronal activation in the forebrain following electrical stimulation of the cuneiform nucleus in the rat: hypothalamic expression of c-fos and NGFI-A messenger RNA. Neuroscience 78(4):1069–1085. Lam W, Verberne AJ (1997) Cuneiform nucleus stimulation-induced sympathoexcitation: role of adrenoceptors, excitatory amino acid and serotonin receptors in rat spinal cord. Brain Res 757(2): 191–201. Mitchell IJ, Dean P, Redgrave P (1988) The projection from superior colliculus to cuneiform area in the rat. II. Defencelike responses to stimulation with glutamate in cuneiform nucleus and surrounding structures. Exp Brain Res 72(3): 626–639. Nag S, Mokha SS (2004) Estrogen attenuates antinociception produced by stimulation of Kolliker-Fuse nucleus in the rat. Eur J Neurosci 20(11):3203–3207. Paxinos G, Watson C (2005) The rat brain in stereotaxic coordinates. London, UK: Elsevier. Richter RC, Behbehani MM (1991) Evidence for glutamic acid as a possible neurotransmitter between the mesencephalic nucleus cuneiformis and the medullary nucleus raphe magnus in the lightly anesthetized rat. Brain Res 544:279–286. Saper CB, Loewy AD (1980) Efferent connections of the parabrachial nucleus in the rat. Brain Res 197(2):291–317. Shafei MN, Nasimi A (2011) Effect of glutamate stimulation of the cuneiform nucleus on cardiovascular regulation in anesthetized rats: Role of the pontine Kolliker-Fuse nucleus. Brain Res 1385: 135–143. Siddall PJ, Polson JW, Dampney RA (1994) Descending antinociceptive pathway from the rostral ventrolateral medulla: a correlative anatomical and physiological study. Brain Res 645(1– 2):61–68. Song G, Xu H, Wang H, Macdonald SM, Poon CS (2011) Hypoxiaexcited neurons in NTS send axonal projections to Kolliker-Fuse/ parabrachial complex in dorsolateral pons. Neuroscience 175:145–153.
446
A. Nasimi et al. / Neuroscience 223 (2012) 439–446
Verberne AJ, Lam W, Owens NC, Sartor D (1997) Supramedullary modulation of sympathetic vasomotor function. Clin Exp Pharmacol Physiol 24(9–10):748–754. Verberne AJ (1995) Cuneiform nucleus stimulation produces activation of medullary sympathoexcitatory neurons in rats. Am J Physiol 268(3 Pt 2):R752–R758.
Zemlan FP, Behbehani MM (1988) Nucleus cuneiformis and pain modulation: anatomy and behavioral pharmacology. Brain Res 453(1–2):89–102.
(Accepted 24 July 2012) (Available online 31 July 2012)