Dysfunction of Mitochondrial Respiratory Chain in Rostral Ventrolateral Medulla During Experimental Endotoxemia Sepsis is a complex pathophysiologic state resulting from an exaggerated whole-body inflammatory response to infection or injury. Metabolic disturbances, abnormal regulation of blood flow and diminished utilization of oxygen at the cellular level may account for tissue damage and lead to multiple organ failure and death. As the primary site of cellular energy generation is the mitochondrion, it presents itself as an important target for the septic cascade. In this regard, the notion that bioenergetic failure due to mitochondrial dysfunction contributes to organ failure during sepsis has received attention. We established the low frequency fluctuations in the systemic arterial pressure signals are related to the sympathetic neurogenic vasomotor tone, and reflect the functional integrity of the brain stem. Their origin is subsequently traced to the premotor sympathetic neurons at the rostral ventrolateral medulla (RVLM), whose neuronal activity is intimately related to the ¡§life-and-death¡¨ process. Based on a rat model of experimental endotoxemia that provides continuous information on changes in neuronal activity in the RVLM, the present study was undertaken to evaluate whether changes in mitochondrial respiratory functions are associated with death arising from sepsis. We also evaluated the efficacy of a new water-soluble coenzyme Q10 (CoQ10, ubiquinone) formula in the protection against fatality during endotoxemia by microinjection into bilateral RVLM. Dysfunction of Mitochondrial Respiratory Chain in Rostral Ventrolateral Medulla During Experimental Endotoxemia in the Rat We investigated the functional changes in mitochondrial respiratory chain at the RVLM in an experimental model of endotoxemia that mimics systemic inflammatory response syndrome. Experiments were carried out in adult male Sprague-Dawley rats that were maintained under propofol anesthesia. Intravenous administration of E. coli lipopolysaccharide (LPS; 30 mg/kg) induced progressive hypotension, with death ensued within 4 hours. The sequence of cardiovascular events during this LPS-induced endotoxemia can be divided into a reduction (Phase I), followed by an augmentation (Phase II; ¡§pro-life¡¨ phase) and a secondary decrease (Phase III; ¡§pro-death¡¨ phase) in the power density of the vasomotor components (0-0.8 Hz) of systemic arterial pressure (SAP) signals. Enzyme assay revealed significant decrease of the activity of NADH cytochrome c reductase (Complex I+III) and cytochrome c oxidase (Complex IV) in the RVLM during all 3 phases of endotoxemia. On the other hand, the activity of succinate cytochrome c reductase (Complex II+III) remained unaltered. Neuroprotective Effects of Coenzyme Q10 at Rostral ventrolateral Medulla Against Fatality During Experimental Endotoxemia in the Rat CoQ10 is a highly mobile electron carrier in the mitochondrial respiratory chain that also acts as an antioxidant. We evaluated the neuroprotective efficacy of CoQ10 against fatality in an experimental model of endotoxemia, using a novel water-soluble formulation of this quinone derivative. In Sprague-Dawley rats maintained under propofol anesthesia, intravenous administration of E. coli LPS (30 mg/kg) induced experimental endotoxemia. Pretreatment by microinjection bilaterally of CoQ10 (1 or 2 mg) into RVLM significantly diminished mortality, prolonged survival time, and reduced the slope or magnitude of the LPS-induced hypotension. CoQ10 pretreatment also significantly prolonged the duration of Phase II endotoxemia and augmented the total power density of the vasomotor components of SAP signals in Phase II endotoxemia. The increase in superoxide anion production induced by LPS at the RVLM during Phases II and III endotoxemia was also significantly blunted. Conclusion The present study revealed that selective dysfunction of respiratory enzyme Complexes I and IV in the mitochondrial respiratory chain at the RVLM is closely associated with fatal endotoxemia. CoQ10 provides neuroprotection against fatality during endotoxemia by acting on the RVLM. We further found that a reduction in superoxide anion produced during endotoxemia at the RVLM may be one of the mechanisms that underlie the elicited neuroprotection of CoQ10. These findings therefore open a new direction for future development of therapeutic strategy in this critical, complicated and highly fatal condition known as sepsis.

Heat shock proteins (HSPs) are abundantly produced in cells that are under stress or injury by acting as a chaperone or promoting folding, unfolding, packing, degradation or denaturing of proteins or peptides. This study evaluated the role of HSP60, HSP70 or HSP90 in the rostral ventrolateral medulla (RVLM), in experimental endotoxemia in the rat. Adult, male Sprague-Dawley rats maintained by i.v. infusion propofol (25 mg/kg/h) were used. During experimental endotoxemia induced by intravenous administration of E. coli lipopolysaccharide (LPS, 30 mg/kg; serotype 0111:B4), the power density of the vasomotor component of systemic arterial pressure (SAP) spectrum underwent a decrease (Phase I), followed by an increase (Phase II; ¡§pro-life¡¨) and a secondary decrease (Phase III; ¡§pro-death¡¨). Western blot analysis revealed that HSP60 expression in the RVLM was significantly increased during Phase II and Phase III endotoxemia; and HSP70 expression was maximally increased during Phase II. HSP90 protein expression in the RVLM was not significantly changed during endotoxemia. We further studied the role of HSP60, HSP70 or HSP90 at the RVLM in experimental endotoxemia by pretreating animals with bilaterally microinjection of an anti-HSP serum (HSPAb, 1:20), normal mouse serum, antisense oligodeoxynucleotide (hsp AODN, 50 pmol), sense oligodeoxynucleotide (hsp SODN) or scrambled AODN (hsp SC). Pretreatment with HSP60Ab or hsp60 AODN resulted in significantly higher mortality, shorter survival time and shorter Phase II duration. In addition, the augmented power density of the vasomotor component of SAP signals during Phase II endotoxemia was significantly reduced. Even more detrimental effects were obtained on local application of HSP70Ab or hsp70 AODN into the RVLM. Pretreatment with HSP90Ab or hsp90 AODN was ineffective. We conclude that the expression of HSP60 and HSP70 in the RVLM may play a ¡§pro-life¡¨ role in fatal experimental endotoxemia; and HSP90 may not be involved.

Obesity afflicts more than 30% of the U.S. population and is a major risk factor for the development of hypertension, type II diabetes, and cardiovascular disease. Studies in humans and animals indicate that obesity is associated with increased sympathetic outflow to the vasculature and kidneys. One mechanism postulated to underlie the increase in sympathetic nerve activity (SNA) in obesity is hyperinsulinemia. Little is known regarding the central circuitry underlying elevated SNA and arterial blood pressure (ABP) during hyperinsulinemia and obesity or if sympathoexcitatory circuits are still responsive to insulin in obesity. Hyperinsulinemic-euglycemic clamps elevate SNA to the hind limb vasculature in lean rodents but obesity is associated with resistance to the peripheral and anorexic effects of insulin. Therefore, the first aim was to determine whether diet-induced obesity causes development of insulin resistance in the central circuits mediating SNA. The sympathoexcitatory response to insulin was still intact in diet-induced obese rats indicating a role for insulin in the elevation in SNA and ABP in obesity. The second aim of this project was to identify the specific receptors in the rostral ventrolateral medulla (RVLM) that mediate the elevated SNA during hyperinsulinemia. The RVLM provides basal sympathetic tone and maintains baseline ABP. Glutamate is the major excitatory neurotransmitter and glutamate receptors of the RVLM are known to mediate multiple forms of hypertension. Blockade of RVLM NMDA-specific glutamatergic receptors reverses the increased lumbar SNA associated with hyperinsulinemia. In contrast, blockade of angiotensin II type 1 or melanocortin receptors in the RVLM had no effect on the sympathoexcitatory response to insulin. The goal of the third aim was to identify the cellular mechanisms within RVLM that mediate the elevated SNA and ABP in diet-induced obesity. Blockade of RVLM glutamate receptors reversed the elevated ABP and lumbar SNA associated with diet-induced obesity while it had no effect on rats on a low fat diet or those resistant to weight gain on the high fat diet. Similar to the findings during hyperinsulinemia, blockade of RVLM angiotensin II type 1 or melanocortin receptors had no effect on lumbar SNA or ABP during diet-induced obesity.

Doctor of Philosophy(PhD) / This thesis addresses the neurochemistry and function of specific nuclei in the autonomic nervous system that are crucial mediators of cardiorespiratory regulation. The primary aim is to build on previous knowledge about muscarinic cholinergic mechanisms within cardiorespiratory nuclei located in the ventrolateral medulla oblongata. The general focus is characterisation of gene expression patterns of specific muscarinic receptor subtypes in central nuclei involved in blood pressure control and respiratory control in normal rats. The findings were subsequently extended by characterisation of muscarinic receptor gene expression patterns in 1) a rat model of abnormal blood pressure control (hypertension) (Chapter 3) 2) a rat model of cholinergic sensitivity (Chapter 5) 3) the rat ventral respiratory group (Chapter 6) The results of a series of related investigations that ensued from the initial aims more finely characterise the neurocircuitry of the ventrolateral medulla, from a specifically cholinoceptive approach. All five muscarinic receptor subtypes are globally expressed in the ventrolateral medulla but only the M2R mRNA was significantly elevated in the VLM of hypertensive animals compared to their normotensive controls and in the VLM of animals displaying cholinergic hypersensitivity compared to their resistant controls. Surprisingly, M2R mRNA is absent in catecholaminergic cell groups but abundant in certain respiratory nuclei. Two smaller projects involving gene expression of other neurotransmitter / neuromodulators expressed in cardiorespiratory nuclei were also completed during my candidature. Firstly, the neurochemical characterisation of enkephalinergic neurons in the RVLM, and their relationship with bulbospinal, catecholaminergic neurons in hypertensive compared to normotensive animals was carried out (Chapter 4). A substantial proportion of sympathoexcitatory neurons located in the RVLM were enkephalinergic in nature. However, there was no significant difference in preproenkephalin expression in the RVLM in hypertensive compared to normotensive animals. Secondly, the identification and distribution of components of the renin-angiotensin aldosterone system (RAAS) within the brainstem, and differences in gene expression levels between hypertensive and normotensive animals was also investigated. The RAAS data was not included in this thesis, since the topic digresses substantially from other chapters and since it is published (Kumar et al., 2006). The mRNA expression aldosterone synthase, mineralocorticoid receptor (MR1), 12-lipoxygenase (12-LO), serum- and glucocorticoid- inducible kinase and K-ras) were found to be present at all rostrocaudal levels of the ventrolateral medulla. Expression of MR1 mRNA was lower in the RVLM of SHR compared with WKY rats and 12-LO mRNA levels were lower in the CVLM in SHR compared with WKY rats. Otherwise, there was no difference in gene expression level, or the method of detection was not sensitive enough to detect differences in low copy transcripts between hypertensive and normotensive animals.

The two recently isolated hypothalamic peptides orexin A and orexin B, also known as hypocretin 1 and 2, are reported to be important signaling molecules in feeding and sleep/wakefulness. Orexin-containing neurons in the lateral hypothalamus project to numerous areas of the rat brain and spinal cord including the intermediolateral cell column (IML) of the thoracolumbar spinal cord. An in vivo and in vitro study was undertaken to evaluate the hypothesis that orexins, acting on sympathetic preganglionic neurons (SPNs) in the rat spinal cord, increase sympathetic outflow. First, orexin A (0.3, 1, and 10 nmol) by intrathecal injection increased mean arterial pressure (MAP) and heart rate (HR) by an average of 5, 18, and 30 mmHg and 10, 42, and 85 beats/min in urethane-anesthetized rats. Intrathecal injection of saline had no significant effects. Orexin B (3 nmol) by intrathecal administration increased MAP and HR by an average of 11 mmHg and 40 beats/min. The pressor effects of orexin A were attenuated by prior intrathecal. injection of orexin A antibodies (1:500 dilution) but not by normal serum albumin. Intravenous administration of the α1-adrenergic receptor antagonist prazosin (0.5 mg/kg) or the β-adrenergic receptor antagonist propranolol (0.5 mg/kg) markedly diminished, respectively, the orexin A-induced increase of MAP and HR. Second, whole cell patch recordings were made from antidromically identified SPNs of spinal cord slices from 12- to 16-day-old rats. Superfusion of orexin A or orexin B (100 or 300 nM) excited 12 of 17 SPNs, as evidenced by a membrane depolarization and/or increase of neuronal discharges. Orexin A- or B-induced depolarizations persisted in TTX (0.5 μM)-containing Krebs solution, indicating that the peptide acted directly on SPNs. Results from our in vivo and in vitro studies together with the previous observation of the presence of orexin A-immunoreactive fibers in the IML suggest that orexins, when released within the IML, augment sympathetic outflow by acting directly on SPNs.

Mevinphos (Mev) is an organophosphate insecticide that acts on the rostral ventrolateral medulla (RVLM), the origin of sympathetic vasomotor tone, to induce cardiovascular responses. This study investigated the role of Trk (tropomyosin-related kinase) (receptor form) or Src (non-receptor form) tyrosine kinase at the RVLM in Mev-induced cardiovascular responses. Bilateral microinjection of Mev (10 nmol) into the RVLM elicited two distinct phases of cardiovascular responses, designated Phase I (sympathoexcitatory) and Phase II (sympathoinhibitory) Mev intoxication. Western blot assay showed that whereas p-Trk490 was increased during Phase I, p-Src416 was increased only during Phase II Mev intoxication. Interestingly, application of a Trk specific inhibitor (K252a; 1 pmol) or Src specific inhibitor (SU6656; 100 pmol) into the bilateral RVLM blunted the Mev-elicited sympathoexcitatory or sympathoinhibitory effect, respectively. Besides, K252a was limited to block NOS I protein expression in the RVLM during Mev intoxication, SU6656 only inhibited NOS II protein expression in the RVLM during Mev intoxication. We conclude that Trk tyrosine kinase (p-Trk490) in the RVLM participates in the Phase I cardiovascular responses during Mev intoxication, Src tyrosine kinase (p-Src416) in the RVLM participates in the Phase II cardiovascular responses associated with Mev intoxication.

Nesse estudo investigamos os efeitos da hipóxia tecidual aguda (HA) sobre as propriedades eletrofisiológicas intrínsecas dos neurônios pré-simpáticos bulboespinhais da área rostro-ventrolateral do bulbo (RVLM) de ratos jovens adultos submetidos previamente à hipóxia crônica intermitente (HCI) e os seus respectivos controle. Para marcarmos os neurônios pré-simpáticos bulboespinhais da RVLM, ratos Wistar jovens (P19-P21) anestesiados com ketamina e xilazina, receberam microinjeções bilaterais de rodamina, um traçador fluorescente retrógrado, na coluna intermediolateral da medula espinhal (T3-T6) e 2 dias após a recuperação da cirurgia, os animais foram submetidos ao protocolo de HCI, enquanto que ratos controle foram mantidos em condições de normóxia, durante 10 dias. No décimo primeiro dia, os ratos foram novamente anestesiados para a remoção do cérebro e as fatias do tronco cerebral contendo neurônios pré-simpáticos com marcação positivas foram registrados. Utilizamos a técnica de whole cell patch-clamp para estudo das propriedades eletrofisiológicas desses neurônios. As propriedades eletrofisiológicas intrínsecas foram analisadas antes e após a HA, a qual foi produzida pela perfusão das fatias do tronco cerebral com uma solução hipóxica (95% N2 + 5% CO2) durante 2 minutos na presença de bloqueadores sinápticos excitatórios e inibitórios. Todos os neurônios pré-simpáticos apresentaram característica intrínseca de autodespolarização e a frequência de disparos basal de potenciais de ação (PAs) desses neurônios de ratos do grupo controle e HCI foram similares [Controle= 5,03 ± 0,4 Hz (n=39) vs HCI= 6,31 ± 0,7 Hz (n=31); p > 0,05]. No grupo controle, a HA não alterou a frequência média de disparos de PAs (BS = 5,03 ± 0,4 Hz vs HA = 5,24 ± 0,3 Hz (n=39); p > 0,05], porém revelou diferentes perfis de disparo de PAs após 2 min de exposição à HA: i) 11 neurônios com aumento na frequência de disparos (BS = 5,1 ± 0,7 Hz vs HA = 7 ± 0,7 Hz; p < 0,05]; ii) 21 neurônios sem alteração na frequência de disparos (BS = 4,8 ± 0,5 Hz vs HA = 5,36 ± 0,6 Hz; p > 0,05] e iii) 7 neurônios com diminuição na frequência de disparos (BS = 7,3 ± 1,1 Hz vs HA = 3,6 ± 0,7 Hz; p < 0,05). No grupo HCI, a HA produziu aumento na frequência média de disparos (BS= 6,31 ± 0,7 Hz vs HA= 7,25 ± 0,8 Hz; n=31 - p < 0,05) e na análise do perfil de disparo de PAs, a HA revelou 2 subpopulações: i) 9 neurônios com aumento na frequência de disparos (BS = 4,7 ± 0,8 Hz vs HA = 8,2 ± 1,4 Hz; p < 0,05) e ii) 22 neurônios sem alteração na frequência de disparos (BS = 7,0 ± 1,0 Hz vs HA = 6,8 ± 1,0 Hz; p > 0,05). Esse estudo nos permitiu revelar diferentes subpopulações de neurônios pré-simpáticos que responderam de forma distintas à HA. Os resultados também sugerem que a HCI teria um efeito pré- condicionante na excitabilidade intrínseca dos neurônios pré-simpáticos em resposta à HA / In this study we evaluated the effects of acute hypoxia (AH) on the intrinsic electrophysiological properties of presympathetic neurons from rostro ventrolateral medulla (RVLM) of juvenile rats exposed to chronic intermittent hypoxia (CIH) or normoxic condition (control group). To label the RVLM bulbospinal presympathetic neurons, young Wistar rats (P 19 - 21) anesthetized with ketamine and xylazine, received bilateral microinjections of a fluorescent retrograde tracer (rhodamine retrobeads) were performed into the intermediolateral column of spinal cord (T3-T6) and two days after recovery of the surgery, the animals were submitted to CIH or normoxic protocol, during 10 days. On the 11th day, under anesthesia, brainstem slices were obtained and only the labeled RVLM presympathetic neurons were recorded, using whole-cell patch-clamp approach to study the electrophysiological properties of these neurons. The intrinsic electrophysiological properties were analyzed before and after AH, which was produced by slice perfusion with hypoxic solution (95% N2 and 5% CO2) during 2 min in the presence of excitatory and inhibitory synaptic antagonists. All recorded RVLM presympathetic neurons presented intrinsic pacemaker activity and the baseline firing frequency of these neurons from control and CIH group were similar [Control= 5,03 ± 0,4 Hz (n=39) vs HCI= 6,31 ± 0,7 Hz (n=31); p > 0,05]. In the control group, AH do not change the firing rate (BS = 5,03 ± 0,4 Hz vs HA = 5,24 ± 0,3 Hz (n=39); p > 0,05), but revealed different pattern of firing frequency after 2 min of AH: i) 11 neurons increased the firing frequency (BS = 4,9 ± 0,9 Hz vs HA = 6,9 ± 1,0 Hz; p < 0,05) ; ii) 21 neurons do not change the firing frequency (BS = 4,8 ± 0,5 Hz vs HA = 5,36 ± 0,6 Hz; p > 0,05) and iii) 7 neurons decreased the firing frequency (BS = 7,3 ± 1,1 Hz vs HA = 3,6 ± 0,7 Hz; p < 0,05). In the CIH group, the AH increased the firing rate comparing with basal condition (SB= 6,31 ± 0,7 Hz vs AH= 7,25 ± 0,8 Hz; n=31 - p < 0,05) and analyzing the pattern of action potential, AH revealed 2 subpopulations in this group: i) 9 neurons increased the firing frequency (SB = 4,7 ± 0,8 Hz vs AH = 8,2 ± 1,4 Hz; p < 0,05) and ii) 22 neurons do not change the firing frequency (SB = 7,0 ± 1,0 Hz vs AH = 6,8 ± 1,0 Hz; p > 0,05).. The data shows that AH revealed different subpopulations of presympathetic neurons and suggest that CIH plays a preconditioning in the intrinsic excitability of presympathetic neurons in response to acute hypoxia

Nesse estudo investigamos os efeitos da hipóxia tecidual aguda (HA) sobre as propriedades eletrofisiológicas intrínsecas dos neurônios pré-simpáticos bulboespinhais da área rostro-ventrolateral do bulbo (RVLM) de ratos jovens adultos submetidos previamente à hipóxia crônica intermitente (HCI) e os seus respectivos controle. Para marcarmos os neurônios pré-simpáticos bulboespinhais da RVLM, ratos Wistar jovens (P19-P21) anestesiados com ketamina e xilazina, receberam microinjeções bilaterais de rodamina, um traçador fluorescente retrógrado, na coluna intermediolateral da medula espinhal (T3-T6) e 2 dias após a recuperação da cirurgia, os animais foram submetidos ao protocolo de HCI, enquanto que ratos controle foram mantidos em condições de normóxia, durante 10 dias. No décimo primeiro dia, os ratos foram novamente anestesiados para a remoção do cérebro e as fatias do tronco cerebral contendo neurônios pré-simpáticos com marcação positivas foram registrados. Utilizamos a técnica de whole cell patch-clamp para estudo das propriedades eletrofisiológicas desses neurônios. As propriedades eletrofisiológicas intrínsecas foram analisadas antes e após a HA, a qual foi produzida pela perfusão das fatias do tronco cerebral com uma solução hipóxica (95% N2 + 5% CO2) durante 2 minutos na presença de bloqueadores sinápticos excitatórios e inibitórios. Todos os neurônios pré-simpáticos apresentaram característica intrínseca de autodespolarização e a frequência de disparos basal de potenciais de ação (PAs) desses neurônios de ratos do grupo controle e HCI foram similares [Controle= 5,03 ± 0,4 Hz (n=39) vs HCI= 6,31 ± 0,7 Hz (n=31); p > 0,05]. No grupo controle, a HA não alterou a frequência média de disparos de PAs (BS = 5,03 ± 0,4 Hz vs HA = 5,24 ± 0,3 Hz (n=39); p > 0,05], porém revelou diferentes perfis de disparo de PAs após 2 min de exposição à HA: i) 11 neurônios com aumento na frequência de disparos (BS = 5,1 ± 0,7 Hz vs HA = 7 ± 0,7 Hz; p < 0,05]; ii) 21 neurônios sem alteração na frequência de disparos (BS = 4,8 ± 0,5 Hz vs HA = 5,36 ± 0,6 Hz; p > 0,05] e iii) 7 neurônios com diminuição na frequência de disparos (BS = 7,3 ± 1,1 Hz vs HA = 3,6 ± 0,7 Hz; p < 0,05). No grupo HCI, a HA produziu aumento na frequência média de disparos (BS= 6,31 ± 0,7 Hz vs HA= 7,25 ± 0,8 Hz; n=31 - p < 0,05) e na análise do perfil de disparo de PAs, a HA revelou 2 subpopulações: i) 9 neurônios com aumento na frequência de disparos (BS = 4,7 ± 0,8 Hz vs HA = 8,2 ± 1,4 Hz; p < 0,05) e ii) 22 neurônios sem alteração na frequência de disparos (BS = 7,0 ± 1,0 Hz vs HA = 6,8 ± 1,0 Hz; p > 0,05). Esse estudo nos permitiu revelar diferentes subpopulações de neurônios pré-simpáticos que responderam de forma distintas à HA. Os resultados também sugerem que a HCI teria um efeito pré- condicionante na excitabilidade intrínseca dos neurônios pré-simpáticos em resposta à HA / In this study we evaluated the effects of acute hypoxia (AH) on the intrinsic electrophysiological properties of presympathetic neurons from rostro ventrolateral medulla (RVLM) of juvenile rats exposed to chronic intermittent hypoxia (CIH) or normoxic condition (control group). To label the RVLM bulbospinal presympathetic neurons, young Wistar rats (P 19 - 21) anesthetized with ketamine and xylazine, received bilateral microinjections of a fluorescent retrograde tracer (rhodamine retrobeads) were performed into the intermediolateral column of spinal cord (T3-T6) and two days after recovery of the surgery, the animals were submitted to CIH or normoxic protocol, during 10 days. On the 11th day, under anesthesia, brainstem slices were obtained and only the labeled RVLM presympathetic neurons were recorded, using whole-cell patch-clamp approach to study the electrophysiological properties of these neurons. The intrinsic electrophysiological properties were analyzed before and after AH, which was produced by slice perfusion with hypoxic solution (95% N2 and 5% CO2) during 2 min in the presence of excitatory and inhibitory synaptic antagonists. All recorded RVLM presympathetic neurons presented intrinsic pacemaker activity and the baseline firing frequency of these neurons from control and CIH group were similar [Control= 5,03 ± 0,4 Hz (n=39) vs HCI= 6,31 ± 0,7 Hz (n=31); p > 0,05]. In the control group, AH do not change the firing rate (BS = 5,03 ± 0,4 Hz vs HA = 5,24 ± 0,3 Hz (n=39); p > 0,05), but revealed different pattern of firing frequency after 2 min of AH: i) 11 neurons increased the firing frequency (BS = 4,9 ± 0,9 Hz vs HA = 6,9 ± 1,0 Hz; p < 0,05) ; ii) 21 neurons do not change the firing frequency (BS = 4,8 ± 0,5 Hz vs HA = 5,36 ± 0,6 Hz; p > 0,05) and iii) 7 neurons decreased the firing frequency (BS = 7,3 ± 1,1 Hz vs HA = 3,6 ± 0,7 Hz; p < 0,05). In the CIH group, the AH increased the firing rate comparing with basal condition (SB= 6,31 ± 0,7 Hz vs AH= 7,25 ± 0,8 Hz; n=31 - p < 0,05) and analyzing the pattern of action potential, AH revealed 2 subpopulations in this group: i) 9 neurons increased the firing frequency (SB = 4,7 ± 0,8 Hz vs AH = 8,2 ± 1,4 Hz; p < 0,05) and ii) 22 neurons do not change the firing frequency (SB = 7,0 ± 1,0 Hz vs AH = 6,8 ± 1,0 Hz; p > 0,05).. The data shows that AH revealed different subpopulations of presympathetic neurons and suggest that CIH plays a preconditioning in the intrinsic excitability of presympathetic neurons in response to acute hypoxia