Central blockade of nitric oxide synthesis reduces moxonidine-induced hypotension

Vestibular neurons with direct projections to the solitary nucleus in the rat

Anterograde and retrograde tract tracing were combined with neurotransmitter and modulator immunolabeling to identify the chemical anatomy of vestibular nuclear neurons with direct projections to the solitary nucleus in rats. Direct, sparsely branched but highly varicose axonal projections from neurons in the caudal vestibular nuclei to the solitary nucleus were observed. The vestibular neurons giving rise to these projections were predominantly located in ipsilateral medial vestibular nucleus. The cell bodies were intensely glutamate immunofluorescent, and their axonal processes contained vesicular glutamate transporter 2, supporting the interpretation that the cells utilize glutamate for neurotransmission. The glutamate-immunofluorescent, retrogradely filled vestibular cells also contained the neuromodulator imidazoleacetic acid ribotide, which is an endogenous CNS ligand that participates in blood pressure regulation. The vestibulo-solitary neurons were encapsulated by axo-somatic GABAergic terminals, suggesting that they are under tight inhibitory control. The results establish a chemoanatomical basis for transient vestibular activation of the output pathways from the caudal and intermediate regions of the solitary nucleus. In this way, changes in static head position and movement of the head in space may directly influence heart rate, blood pressure, respiration, as well as gastrointestinal motility. This would provide one anatomical explanation for the synchronous heart rate and blood pressure responses observed after peripheral vestibular activation, as well as disorders ranging from neurogenic orthostatic hypotension, postural orthostatic tachycardia syndrome, and vasovagal syncope to the nausea and vomiting associated with motion sickness.NEW & NOTEWORTHY Vestibular neurons with direct projections to the solitary nucleus utilize glutamate for neurotransmission, modulated by imidazoleacetic acid ribotide. This is the first direct demonstration of the chemical neuroanatomy of the vestibulo-solitary pathway.

Imidazoleacetic acid-ribotide in vestibulo-sympathetic pathway neurons

Imidazole-4-acetic acid-ribotide (IAARP) is a putative neurotransmitter/modulator and an endogenous regulator of sympathetic drive, notably systemic blood pressure, through binding to imidazoline receptors. IAARP is present in neurons and processes throughout the CNS, but is particularly prevalent in regions that are involved in blood pressure control. The goal of this study was to determine whether IAARP is present in neurons in the caudal vestibular nuclei that participate in the vestibulo-sympathetic reflex (VSR) pathway. This pathway is important in modulating blood pressure upon changes in head position with regard to gravity, as occurs when humans rise from a supine position and when quadrupeds climb or rear. Sinusoidal galvanic vestibular stimulation was used to activate the VSR and cfos gene expression in VSR pathway neurons of rats. These subjects had previously received a unilateral FluoroGold tracer injection in the rostral or caudal ventrolateral medullary region. The tracer was transported retrogradely and filled vestibular neuronal somata with direct projections to the injected region. Brainstem sections through the caudal vestibular nuclei were immunostained to visualize FluoroGold, cFos protein, IAARP and glutamate immunofluorescence. The results demonstrate that IAARP is present in vestibular neurons of the VSR pathway, where it often co-localizes with intense glutamate immunofluorescence. The co-localization of IAARP and intense glutamate immunofluorescence in VSR neurons may represent an efficient chemoanatomical configuration, allowing the vestibular system to rapidly up- and down-modulate the activity of presympathetic neurons in the ventrolateral medulla, thereby altering blood pressure.

Crosstalk between central pathways of nitric oxide and carbon monoxide in the hypertensive action of cyclosporine

Although the intermediary role of central neurons in the hypertensive and sympathoexcitatory actions of cyclosporine (CSA) has been recognized in previous studies including our own, the underlying mechanism remains obscure. In this study, we tested the hypothesis that central pathways of nitric oxide (NO) and carbon monoxide (CO) modulate the blood pressure (BP) response elicited by CSA in conscious rats. Hemodynamic effects of CSA were evaluated in absence and presence of maneuvers that inhibit or facilitate biosynthesizing enzymes of NO (NOS) or CO (heme oxygenase, HO). CSA (20mg/kg i.v.) produced abrupt increases in BP that peaked in 5min and maintained for at least 45min. The hypertensive effect of CSA disappeared in rats pretreated intracisternally (i.c.) with N(ω)-nitro-l-arginine methyl ester (L-NAME, nonselective NOS inhibitor), N(5)-(1-iminoethyl)-l-ornithine (L-NIO, selective eNOS inhibitor), N(ω)-propyl-l-arginine (NPLA, selective nNOS inhibitor), or 1H-[1,2,4] oxadiazolo[4,3-a] quinoxalin-1-one (ODQ, guanylate cyclase inhibitor), suggesting the importance of central eNOS/nNOS/GC cascade in CSA-induced hypertension. L-NAME also abolished the hypotension caused by the sympatholytic drug moxonidine, indicating a tonic sympathoinhibitory action for NO. The inhibition of HO activity by zinc protoporphyrin IX (ZnPP) abrogated the hypertensive action of CSA. The abolition by L-NAME or ZnPP of CSA hypertension was compromised upon simultaneous i.c. exposure to hemin (HO substrate) and l-arginine (NOS substrate), respectively. Together, the interruption of the mutually facilitated NOS/NO and HO/CO pathways and coupled GC/cGMP in central neuronal pools accounts, at least partly, for the hypertensive and perhaps sympathoexcitatory actions of CSA.

Anatomical observations of the caudal vestibulo-sympathetic pathway

The vestibular system senses the movement and position of the head in space and uses this information to stabilize vision, control posture, perceive head orientation and self-motion in three-dimensional space, and modulate autonomic and limbic activity in response to locomotion and changes in posture. Most vestibular signals are not consciously perceived and are usually appreciated through effector pathways classically described as the vestibulo-ocular, vestibulo-spinal, vestibulo-collic and vestibulo-autonomic reflexes. The present study reviews some of the recent data concerning the connectivity and chemical anatomy of vestibular projections to autonomic sites that are important in the sympathetic control of blood pressure.

GABAergic mechanism in the rostral ventrolateral medulla contributes to the hypotension of moxonidine

AIMS

The depressor action of the centrally antihypertensive drug moxonidine has been attributed to activation of I(1)-imidazoline receptor in the rostral ventrolateral medulla (RVLM). The objective of this study was to determine the role of the γ-aminobutyric acid (GABA) mechanisms in the RVLM in mediating the effect of moxonidine in anaesthetized normotensive rats.

METHODS AND RESULTS

The relationship between the effects of microinjection or picoinjection of moxonidine and the functional state of GABA receptors at the level of the RVLM or pre-sympathetic neuron was determined. Microdialysis was performed to detect the effect of moxonidine on the release of GABA in the RVLM. Western blot analysis was carried out to test the effect of chronic intracerebroventricular injection of moxonidine on the protein expression of GABA receptors in the RVLM. Pre-treatment with the GABA(A) or GABA(B) receptor antagonist bicuculline (5 pmol) or CGP35348 (200 pmol), respectively, microinjected into the RVLM significantly attenuated the decrease in blood pressure and renal sympathetic nerve activity induced by moxonidine. In 22 moxonidine-sensitive pre-sympathetic neurons in the RVLM, picoinjection of bicuculline (100 fmol/5 nL) significantly attenuated the neuronal inhibition evoked by moxonidine (100 pmol/5 nL). The release of GABA in the RVLM was increased after intravenous moxonidine (50 μg/kg). Central infusion of moxonidine upregulated the protein expression of both GABA(A) and GABA(B) receptors in the RVLM.

CONCLUSION

The current data demonstrate that GABAergic mechanisms in the RVLM are responsible for the hypotension and sympathoinhibition of moxonidine.

Differential central NOS-NO signaling underlies clonidine exacerbation of ethanol-evoked behavioral impairment

BACKGROUND

The molecular mechanisms that underlie clonidine exacerbation of behavioral impairment caused by ethanol are not fully known. We tested the hypothesis that nitric oxide synthase (NOS)-derived nitric oxide (NO) signaling in the locus coeruleus (LC) is implicated in this phenomenon.

METHODS

Male Sprague-Dawley rats with intracisternal (i.c.) and jugular vein cannulae implanted 6 days earlier were tested for drug-induced behavioral impairment. The latter was assessed as the duration of loss of righting reflex (LORR) and rotorod performance every 15 minutes until the rat recovered to the baseline walk criterion (180 seconds). In a separate cohort, we measured p-neuronal NOS (nNOS), p-endothelial NOS (eNOS), and p-ERK1/2 in the LC following drug treatment, vehicle, or NOS inhibitor.

RESULTS

Rats that received clonidine [60 Ig/kg, i.v. (intravenous)] followed by ethanol (1 or 1.5 g/kg, i.v.) exhibited synergistic impairment of rotorod performance. Intracisternal pretreatment with nonselective NOS inhibitor N(omega)-nitro-L-arginine methyl ester (L-NAME, 0.5 mg) or selective nNOS inhibitor N-propyl-L-arginine (1 microg) exacerbated the impairment of rotorod performance caused by clonidine-ethanol combination. Exacerbation of behavioral impairment was caused by L-NAME enhancement of the effect of ethanol, not clonidine. L-NAME did not influence blood ethanol levels; thus, the interaction was pharmacodynamic. LORR caused by clonidine (60 microg/kg, i.v.)-ethanol (1 g/kg, i.v.) combination was abolished by selective inhibition of central eNOS (L-NIO, 10 microg i.c.) but not by nNOS inhibition under the same conditions. Western blot analyses complemented the pharmacological evidence by demonstrating that clonidine-ethanol combination inhibits phosphorylation (activation) of nNOS (p-nNOS) and increases the level of phosphorylated eNOS (p-eNOS) in the LC; the change in p-nNOS was paralleled by similar change in LC p-ERK1/2. NOS inhibitors alone did not affect the level of nitrate/nitrite, p-nNOS, p-eNOS, or p-ERK1/2 in the LC.

CONCLUSIONS

Alterations in NOS-derived NO in the LC underlie clonidine-ethanol induced behavioral impairment. A decrease in nNOS activity, due at least partly to a reduction in nNOS phosphorylation, mediates rotorod impairment, while enhanced eNOS activity contributes to LORR, elicited by clonidine-ethanol combination.

Facilitation of central imidazoline I(1)-site/extracellular signal-regulated kinase/p38 mitogen-activated protein kinase signalling mediates the hypotensive effect of ethanol in rats with acute renal failure

BACKGROUND AND PURPOSE

This study investigated the role of central sympathetic activity and related mitogen-activated protein kinase (MAPK) signalling in the cardiovascular effects of ethanol in a model of acute renal failure (ARF).

EXPERIMENTAL APPROACH

The effects of pharmacological interventions that inhibit peripheral or central sympathetic activity or MAPK on the cardiovascular actions of ethanol in rats with ARF induced by glycerol were evaluated.

KEY RESULTS

Glycerol (50%, 10 mL.kg(-1), i.m.) caused progressive increases and decreases in blood pressure (BP) and heart rate (HR) respectively. Subsequent i.v. ethanol (0.25 or 1 g.kg(-1)) elicited dose-related changes in BP (decreases) and HR (increases). These effects were replicated after intracisternal (i.c.) administration of ethanol. Blockade of nicotinic cholinoceptors (nAChR, hexamethonium, 20 mg.kg(-1)) or alpha(1)-adrenoceptors (prazosin, 1 mg.kg(-1)) attenuated cardiovascular effects of ethanol. Ethanol hypotension was also attenuated after the centrally acting sympatholytic drug moxonidine (selective I(1)-site agonist, 100 microg.kg(-1) i.v.), but not guanabenz (selective alpha(2)-receptor agonist, 30 microg.kg(-1), i.v.), suggesting involvement of central circuits of I(1) sites in ethanol-evoked hypotension. Selective blockade I(1) sites (efaroxan) but not alpha(2) (yohimbine) adrenoceptors abolished the hypotensive response to ethanol. Intracisternal administration of PD98059 or SB203580, inhibitors of extracellular signal-regulated kinase (ERK 1/2) and p38 MAPK, respectively, reduced the hypotensive action of moxonidine or ethanol. When used simultaneously, the two MAPK inhibitors produced additive attenuation of ethanol hypotension.

CONCLUSIONS AND IMPLICATIONS

Sympathoinhibitory pathways of central I(1)-sites and downstream ERK/p38 MAPK signalling were involved in the hypotensive action of ethanol in ARF.

Sympathoinhibitory mechanism of moxonidine: role of the inducible nitric oxide synthase in the rostral ventrolateral medulla

AIMS

The central antihypertensive drug moxonidine lowers blood pressure (BP) through stimulating an imidazoline receptor within the rostral ventrolateral medulla (RVLM). Nitric oxide (NO) generated by the inducible NO synthase (iNOS) in the RVLM has been suggested to be involved in tonic sympathetic inhibition. The aim of this study was to determine the role of NO generated by iNOS in mediating moxonidine-induced cardiovascular inhibition in rats.

METHODS AND RESULTS

In anaesthetized rats, the cardiovascular response to local or systemic injection of moxonidine was observed after treatment with the selective iNOS inhibitor S-methylisothiourea (SMT) in the brain. Using immunohistochemical staining and western blot techniques, the protein expression of iNOS in the RVLM was measured in the moxonidine-infused rats. Intracerebroventricular (ICV) injection of SMT (1-100 nmol) dose-dependently attenuated the moxonidine (20 nmol, ICV)-induced decrease in BP and heart rate. Prior injection of SMT (20 and 200 pmol) into the RVLM also dose-dependently prevented the decrease in BP and renal sympathetic nerve activity evoked by RVLM microinjection of moxonidine (5 nmol) or intravenous injection of moxonidine (50 microg/kg). We further found that expression of iNOS protein following chronic ICV infusion of moxonidine (20 nmol, 2 weeks) is selectively upregulated in the RVLM but not in the nucleus tractus solitarius.

CONCLUSION

The present data suggest that an NO mechanism generated by iNOS in the RVLM plays an important role in mediating the sympathetic inhibition of the centrally acting drug moxonidine.

Control of left ventricular mass by moxonidine involves reduced DNA synthesis and enhanced DNA fragmentation

BACKGROUND AND PURPOSE

Left ventricular hypertrophy (LVH) is a maladaptive process associated with increased cardiovascular risk. Regression of LVH is associated with reduced complications of hypertension. Moxonidine is an antihypertensive imidazoline compound that reduces blood pressure primarily by central inhibition of sympathetic outflow and by direct actions on the heart to release atrial natriuretic peptide, a vasodilator and an antihypertrophic cardiac hormone. This study investigated the effect of moxonidine on LVH and the mechanisms involved in this effect.

EXPERIMENTAL APPROACH

Spontaneously hypertensive rats were treated with several doses of moxonidine (s.c.) over 4 weeks. Blood pressure and heart rate were continuously monitored by telemetry. Body weight and water and food intake were measured weekly. Measurements also included left ventricular mass, DNA content, synthesis, fragmentation, and apoptotic/anti-apoptotic pathway proteins.

KEY RESULTS

The decrease in mean arterial pressure stabilized at approximately -10 mm Hg after 1 week of treatment and thereafter. Compared to vehicle-treated rats (100%), left ventricular mass was dose- and time-dependently reduced by treatment. This reduction remained significantly lower after normalizing to body weight. Moxonidine reduced left ventricular DNA content and inhibited DNA synthesis. DNA fragmentation transiently, but significantly increased at 1 week of moxonidine treatment and was paralleled by elevated active caspase-3 protein. The highest dose significantly decreased the apoptotic protein Bax and all doses stimulated anti-apoptotic Bcl-2 after 4 weeks of treatment.

CONCLUSIONS AND IMPLICATIONS

These studies implicate the modulation of cardiac DNA dynamics in the control of left ventricular mass by moxonidine in a rat model of hypertension.

Brainstem phosphorylated extracellular signal-regulated kinase 1/2-nitric-oxide synthase signaling mediates the adenosine A2A-dependent hypotensive action of clonidine in conscious aortic barodenervated rats

The cellular mechanisms that underlie the enhancement of clonidine-evoked hypotension in aortic barodenervated (ABD) rats and its dependence on central adenosine A(2A) receptor (A(2A)R) are not known. We tested the hypothesis that A(2A)R-mediated phosphorylation of extracellular signal-regulated kinase (pERK)1/2 in the rostral ventrolateral medulla (RVLM) and its downstream activation of nitric-oxide synthase (NOS)-NO signaling underlie the centrally (clonidine)-mediated hypotension. We first demonstrated an up-regulation of the molecular targets for clonidine [imidazoline I(1) and alpha(2A) adrenergic receptors (alpha(2A)R)] in the RVLM of ABD compared with sham-operated (SO) rats; this finding might explain the enhanced clonidine hypotension in ABD rats. A similar anatomical up-regulation of the RVLM A(2A)R was evident and was complemented with enhanced central A(2A)R [2-[4-[(2-carboxyethyl)phenyl]ethylamino]-5'-N-ethylcarboxamidoadenosine; CGS21680]-mediated hypotension in ABD rats. The hypotension produced by intracisternal CGS21680 or clonidine, in conscious ABD rats, was associated with a significant increase in pERK1/2 level in the RVLM. Whereas selective A(2A)R blockade [5-amino-7-(2-phenylethyl)-2-(2-furyl)-pyrazolo[4,3-epsilon]-1,2,4-triazolo[1,5-c]pyrimidine; SCH58261] or NOS inhibition (N(omega)-nitro-l-arginine methyl ester) virtually abolished clonidine-evoked hypotension, clonidine-evoked enhancement of RVLM pERK1/2 production was only abrogated by SCH58261 pretreatment. These findings suggest that interventions that act centrally to increase RVLM neuronal pERK1/2 production elicit hypotension via the activation of downstream NOS-NO signaling. The findings also yield insight into a cellular mechanism that might explain the dependence of centrally (clonidine)-mediated hypotension on central A(2A)R signaling in the ABD rat.

Involvement of central α1- and α2-adrenoceptors on cardiovascular responses to moxonidine

In the present study we compared the effects produced by moxonidine (alpha2-adrenoceptor/imidazoline agonist) injected into the 4th cerebral ventricle and into the lateral cerebral ventricle on mean arterial pressure, heart rate and on renal, mesenteric and hindquarter vascular resistances, as well as the possible action of moxonidine on central alpha1- or alpha2-adrenoceptors to produce cardiovascular responses. Male Holtzman rats (n=7-8) anesthetized with urethane (0.5 g/kg, intravenously-i.v.) and alpha-chloralose (60 mg/kg, i.v.) were used. Moxonidine (5, 10 and 20 nmol) injected into the 4th ventricle reduced arterial pressure (-19+/-5, -30+/-7 and -43+/-8 mmHg vs. vehicle: 2+/-4 mmHg), heart rate (-10+/-6, -16+/-7 and -27+/-9 beats per minute-bpm, vs. vehicle: 4+/-5 bpm), and renal, mesenteric and hindquarter vascular resistances. Moxonidine (5, 10 and 20 nmol) into the lateral ventricle only reduced renal vascular resistance (-77+/-17%, -85+/-13%, -89+/-10% vs. vehicle: 3+/-4%), without changes on arterial pressure, heart rate and mesenteric and hindquarter vascular resistances. Pre-treatment with the selective alpha2-adrenoceptor antagonist yohimbine (80, 160 and 320 nmol) injected into the 4th ventricle attenuated the hypotension (-32+/-5, -25+/-4 and -12+/-6 mmHg), bradycardia (-26+/-11, -23+/-5 and -11+/-6 bpm) and the reduction in renal, mesenteric and hindquarter vascular resistances produced by moxonidine (20 nmol) into the 4th ventricle. Pre-treatment with yohimbine (320 nmol) into the lateral ventricle did not change the renal vasodilation produced by moxonidine (20 nmol) into the lateral ventricle. The alpha1-adrenoceptor antagonist prazosin (320 nmol) injected into the 4th ventricle did not affect the cardiovascular effects of moxonidine. However, prazosin (80, 160 and 320 nmol) into the lateral ventricle abolished the renal vasodilation (-17+/-4, -6+/-9 and 2+/-11%) produced by moxonidine. The results indicate that the decrease in renal vascular resistance due to moxonidine action in the forebrain is mediated by alpha1-adrenoceptors, while the cardiovascular effects produced by moxonidine acting in the brainstem depend at least partially on the activation of alpha2-adrenoceptors.

Vestibular neurons in the rat contain imidazoleacetic acid-ribotide, a putative neurotransmitter involved in blood pressure regulation

A substantial body of research has led to the recognition that the vestibular system participates in blood pressure modulation during active movements and changes in posture, and that this modulation is effected at least partly by the caudal vestibular nuclei. The I-4 isomer of imidazoleacetic acid-ribotide (IAA-RP) is a putative neurotransmitter/modulator that is thought to be an endogenous regulator of general sympathetic drive, particularly systemic blood pressure. The present study employed immunofluorescence and light and electron microscopic immunocytochemistry to visualize IAA-RP in the vestibular nuclei of adult male rats. The results demonstrate IAA-RP immunolabeling of subpopulations of vestibular neurons in the descending nucleus and the caudal half of the medial nucleus, with scattered immunostained vestibular neurons also present more rostrally. On the basis of double immunofluorescence staining for IAA-RP and calbindin, many of these ribotide-immunoreactive neurons appear to be innervated by cerebellar Purkinje cell afferents. Ultrastructural observations in the caudal vestibular nuclei confirm the IAA-RP immunolocalization in cell bodies and dendritic processes, and in some myelinated axons and presynaptic boutons. The regional distribution of IAA-RP immunoreactivity corresponds to the location of vestibular neurons involved in autonomic functions. The presence of IAA-RP in those neurons suggests that they participate specifically in vestibulo-autonomic regulation of blood pressure. The localization of immunostain in processes and terminals suggests that vestibulo-autonomic activity is subject to local feedback control. Overall, the observations offer a chemoanatomic basis for understanding the vestibular side effects commonly experienced by patients treated with clonidine and other imidazoline-related drugs.

Moxonidine: a review of its use in essential hypertension

Moxonidine (Physiotens, Moxon, Cynt) is an orally administered imidazoline compound with selective agonist activity at imidazoline I1 receptors and only minor activity at alpha2-adrenoceptors. Moxonidine acts centrally to reduce peripheral sympathetic activity, thus decreasing peripheral vascular resistance. In patients with mild to moderate hypertension, moxonidine reduces blood pressure (BP) as effectively as most first-line antihypertensives when used as monotherapy and is also an effective adjunctive therapy in combination with other antihypertensive agents. It improves the metabolic profile in patients with hypertension and diabetes mellitus or impaired glucose tolerance, is well tolerated, has a low potential for drug interactions and may be administered once daily in most patients. Thus, moxonidine is a good option in the treatment of patients with mild to moderate hypertension, particularly as adjunctive therapy in patients with the metabolic syndrome.

Central Adenosine Signaling Plays a Key Role in Centrally Mediated Hypotension in Conscious Aortic Barodenervated Rats

We tested the hypothesis that clonidine-evoked hypotension is dependent on central adenosinergic pathways. Five groups of male, conscious, aortic baroreceptor-denervated (ABD) rats received clonidine (10 microg/kg i.v.) 30 min after i.v. 1) saline, 2) theophylline (10 mg/kg), or 3) 8-(p-sulfophenyl)theophylline (8-SPT) (2.5 mg/kg) or 1 h after i.p. 4) dipyridamole (5 mg/kg) or 5) an equal volume of sesame oil. Blockade of central (theophylline) but not peripheral (8-SPT) adenosine receptors abolished clonidine hypotension. In contrast, dipyridamole substantially enhanced the bradycardic response to clonidine. In additional groups, intracisternal (i.c.) dipyridamole (150 microg) and 8-SPT (10 microg) enhanced and abolished, respectively, clonidine (0.6 microg i.c.)-evoked hypotension. Because clonidine is a mixed I1/alpha2 agonist, we also investigated whether adenosine signaling is linked to the I1 or the alpha2A receptor by administering the selective I1 (rilmenidine, 25 microg) or alpha2A [alpha-methylnorepinephrine (alpha-MNE), 4 microg] agonist 30 min after central adenosine receptor blockade (8-SPT; 10 microg i.c.) or artificial cerebrospinal fluid. The hypotensive response elicited by rilmenidine or alpha-MNE was abolished in 8-SPT-pretreated rats. To delineate the role of the adenosine A2A receptor in clonidine-evoked hypotension, i.c. clonidine (0.6 microg) was administered 30 min after central adenosine receptor A2A blockade [5-amino-7-(2-phenylethyl)-2-(2-furyl)-pyrazolo[4,3-epsilon]-1,2,4-triazolo[1,5-c]-pyrimidine (SCH58261); 150 microg i.c.]. The latter virtually abolished the hypotensive and bradycardic responses elicited by clonidine. In conclusion, central adenosine A2A signaling plays a key role in clonidine-evoked hypotension in conscious aortic barodenervated rats.