Prediction of somatic and non-somatic depressive symptoms between inpatient rehabilitation and follow-up.

STUDY DESIGN: Longitudinal. OBJECTIVE: We identified changes in the association of somatic and non-somatic symptoms (as measured by the Patient Health Questionnaire-9, PHQ-9) between inpatient rehabilitation after spinal cord injury (SCI) and 1 year after discharge. SETTING: A specialty hospital in the Southeastern USA. METHODS: A total of 584 adults with traumatic SCI were administered the PHQ-9 during inpatient rehabilitation. Of them, 227 completed the PHQ-9 by survey at 1-year follow-up. We performed time-lagged regression between times of measurement for somatic and non-somatic factors of the PHQ-9. RESULTS: The non-somatic factor at baseline was significantly predictive of the non-somatic (r=0.67, P=0.002) and somatic factors at follow-up (r=0.53, P=0.019). The somatic factor did not significantly predict either the somatic (r=0.10, n.s.) or non-somatic factors at follow-up (r=-0.01, NS). Factor analysis also indicated changing factor structure between inpatient rehabilitation and follow-up. CONCLUSIONS: Our results question the interpretation of somatic items during inpatient rehabilitation, as they are not predictive of either somatic or non-somatic symptoms at follow-up.

Synaptic abnormalities of mice lacking toll-like receptor (TLR)-9.

Motor, sensory, and autonomic abnormalities are reported for toll-like receptor 9 (TLR9) knock-out (KO) mice. However, a physiological role of TLR9 in the nervous system is largely unknown. Since altered synaptic transmission can contribute to sensory and motor abnormalities, we evaluated neuromuscular junction (NMJ) function and morphology of TLR9 KO mice. Triangularis sterni nerve-muscle preparations were dissected from TLR9 KO and age-matched control mice. Two-electrode voltage clamp of the motor endplate revealed that the amplitude and frequency of miniature end plate currents (mEPCs) for TLR9 KO NMJs were significantly greater than control. In contrast, mean endplate current (EPC, 1Hz) amplitude was equivalent to control. The ratio of mean EPC to mean mEPC amplitude indicated a decline of quantal content (m) for TLR9 KO NMJs. Furthermore, m declined more rapidly than control in response to 50-Hz stimulus trains. A rightward shift of the mEPC amplitude distribution suggested formation of vesicles containing larger amounts of acetylcholine (ACh). Staining with rhodamine α-bungarotoxin revealed a significant decline of endplate size in TLR9 KO mice. This alteration may result from ACh-induced decline of acetylcholine receptor (AChR) expression resulting from increased frequency and amplitude of mEPCs. At the same time, excessive spontaneous vesicular ACh release may initiate retrograde suppression of excitation-secretion coupling. These data suggest a novel role of TLR9 in the development of the NMJ.

Convergence of pre- and postsynaptic influences on glucosensing neurons in the ventromedial hypothalamic nucleus.

Glucosensing neurons in the ventromedial hypothalamic nucleus (VMN) were studied using visually guided slice-patch recording techniques in brain slices from 14- to 21-day-old male Sprague-Dawley rats. Whole-cell current-clamp recordings were made as extracellular glucose levels were increased (from 2.5 to 5 or 10 mmol/l) or decreased (from 2.5 to 0.1 mmol/l). Using these physiological conditions to define glucosensing neurons, two subtypes of VMN glucosensing neurons were directly responsive to alterations in extracellular glucose levels. Another three subtypes were not directly glucose-sensing themselves, but rather were presynaptically modulated by changes in extracellular glucose. Of the VMN neurons, 14% were directly inhibited by decreases in extracellular glucose (glucose-excited [GE]), and 3% were directly excited by decreases in extracellular glucose (glucose-inhibited [GI]). An additional 14% were presynaptically excited by decreased glucose (PED neurons). The other two subtypes of glucosensing neurons were either presynaptically inhibited (PIR; 11%) or excited (PER; 8%) when extracellular glucose was raised to > 2.5 mmol/l. GE neurons sensed decreased glucose via an ATP-sensitive K(+) (K(ATP)) channel. The inhibitory effect of increased glucose on PIR neurons appears to be mediated by a presynaptic gamma-aminobutyric acid-ergic glucosensing neuron that probably originates outside the VMN. Finally, all types of glucosensing neurons were both fewer in number and showed abnormal responses to glucose in a rodent model of diet-induced obesity and type 2 diabetes.

A study of the reinnervation of fast and slow mammalian muscles.

Miniature end plate potential (mepp) frequency in innervated extensor muscle is significantly higher than in soleus muscle. 9 days after nerve crush mepps of low amplitude and prolonged duration reappeared at a frequency of 2% of control and were similar to normal muscles after 35 days. Membrane potential began to increase 9-10 days after nerve crush and at 30 days was similar to controls. The region most sensitive to ACh in denervated and reinnervated muscles was the end plate. Caffeine (20 mM, 23 degrees C) induced contracture in innervated soleus but not in extensor muscles. After denervation the extensor became sensitive to caffeine while the soleus muscles decreased in sensitivity to the drug; 4-5 days after reinnervation the effect of caffeine on these muscles was similar to control. The events during reinnervation are: (a) reappearance of mepps at the same time as end plate potential and muscle twitch; (b) partial restoration of the membrane potential; (c) return of caffeine-induced contracture to normal levels in the soleus and its absence in the extensor muscles; (d) return of membrane resistance to normal values in both muscles at about 25 days; and (e) return of ACh-sensitivity to control levels at about 30 days in both muscles. Although these results suggest that the membrane potential and sarcoplasmic reticulum are under neural influence, it remains to be established whether or not separate neurotrophic factors are involved.

Action potentials in fast- and slow-twitch mammalian muscles during reinnervation and development.

Action potentials (APs) were recorded from the extrajunctional membrane of surface fibers of the fast-twitch extensor digitorum longus (extensor) and the slow-twitch soleus muscles of adult rats. APs of the extensor muscle had a significantly faster rate of rise and fall, as well as a shorter duration, than those of the soleus. In addition, the overshoot of APs and the resting membrane potential was greater for the extensor. Whereas the soleus produced only one AP regardless of the stimulus duration, the number of extensor responses was directly proportional to the stimulus duration. This repetitive activity was greatly reduced by a concentration of tetrodotoxin (TTX) as low as 5 X 10(11) g/ml. Within 8 d after crush of the nerves to these two muscles, all differences in AP properties disappeared and both muscles became partially resistant to TTX. Reinnervation brought about a redifferentiation so that differences in AP were again significant at 22 d after nerve crush. However, the rate of rise of extensor APs did not attain normal values even as late as 60 d after nerve crush. APs were found to be the same for extensor and soleus muscles from 12-d-old rats. At 18 d after birth, rate of rise was equivalent to that of adult muscle for the soleus although 50--60 d were required before this parameter was fully mature for the extensor. Nevertheless, APs of the extensor and soleus were clearly differentiated within 25 d after birth. Differences in fast and slow muscle APs are discussed with regard to differences in ion gradients and sarcolemmal conductance.

Excitatory amino acid induced currents of isolated murine hypothalamic neurons and their suppression by 2,3-butanedione monoxime.

Ionic currents induced by excitatory amino acids were investigated for freshly isolated murine hypothalamic neurons with whole cell recording techniques. L-glutamate or N-methyl-D-aspartate (NMDA), in combination with glycine, resulted in a rapidly rising current which decayed in the continued presence of agonist. In contrast, kainate currents did not decay. While quisqualate-induced current maintained a steady amplitude in the continued presence of agonist, a rapid decay phase appeared at holding potentials negative to -50 mV. Co-application of 2,3-butanedione monoxime (BDM) reversibly inhibited the currents due to each agonist. Detailed study of BDM suppression of kainate-induced current revealed two components. A component with a rapid onset did not involve phosphatase action since 500 microM ATP-gamma-S or a protein kinase inhibitor (H-7, 200 microM) did not alter current suppression or recovery after BDM. Thus, the probable mechanism for this component of BDM's effect is direct block of the kainate-activated ion channel. However, preincubating neurons with 30 mM BDM reduced their subsequent response to kainate alone. This persistent effect of BDM was not seen for neurons dialyzed with a solution containing ATP-gamma-S during conventional whole cell recording. Furthermore, exposure to H-7 prevented recovery of the kainate response suppressed by preincubation in BDM. These findings suggest that BDM causes sustained suppression of the kainate response of hypothalamic neurons via a "chemical phosphatase" action.

Decay of ethanol-induced suppression of glycine-activated current of ventral tegmental area neurons.

We demonstrated previously that ethanol depresses glycine-induced currents in 45% of neurons freshly isolated from the ventral tegmental area (VTA) of rats (), and that protein kinase C (PKC) modulates this action of ethanol (). In the present study, we investigated the time course of this effect of ethanol on VTA neurons from young rats. For 70% of the neurons in which ethanol reduced glycine-evoked currents, this depressant effect gradually diminished during continuous superfusion with ethanol. Its action decayed faster when ethanol was applied in several brief pulses than by continuous superfusion. On the other hand, the decay was especially slower when ethanol was applied in pulses at longer intervals or by preincubation. Phorbol ester 12,13-dibutyrate (PDBu, 1 microM), an activator of PKC, also depressed glycine-induced currents. In approximately 40% (6/15) of the neurons, the effect of PDBu diminished with time and was antagonized by the specific PKC inhibitor, chelerythrine (7 microM). Chelerythrine also attenuated the ethanol-induced depression of glycine-induced currents and its time-dependent decay, thus confirming our previous evidence that PKC mediates, at least in part, the decay of the depressant effect of ethanol on glycine-induced currents of VTA neurons.

Elevated density and altered pharmacologic properties of myocardial calcium current of the spontaneously hypertensive rat.

DESIGN: The membrane current mediated by the L-type calcium channel (ICa) was studied for myocytes isolated from the ventricle of 10- to 11-week-old spontaneously hypertensive rats (SHR). RESULTS: Compared with age-matched normotensive Wistar-Kyoto (WKY) or Sprague-Dawley rats, the amplitude of ICa was greater for the SHR. Two observations suggest that the greater ICa of the SHR was not due to hypertrophy. First, the similarity of membrane capacitance for these three strains of rat indicated lack of hypertrophy. Secondly, the amplitude of ICa was also greater for myocytes isolated from the right ventricle of the SHR. The ICa of the SHR was more sensitive to drugs known to activate calcium channels via phosphorylation. Specifically, extracellular application of 1 mumol/l isoprenaline as well as intracellular dialysis with either 1 mmol/l cyclic AMP or with 1 mmol/l adenosine 5'-O-3-thiotriphosphate increased the mean ICa of SHR myocytes significantly more than that of WKY rat cells. The ICa of SHR myocytes was also more sensitive to BAY K 8644 and its enantiomorphs. CONCLUSION: The present data suggest that the greater peak amplitude of ICa for SHR myocytes relative to that of myocytes of normotensive rats is due to an increase in current density and enhancement of channel phosphorylation.

Effects of butanedione monoxime on neuromuscular transmission.

1. The amplitude of endplate potentials was increased by concentrations of butanedione monoxime (BDM, 5-20 mM) that typically caused muscle paralysis. 2. Although BDM slowed the decay of spontaneous miniature endplate currents, the effect was insufficient to explain most of the large increase in amplitude of endplate potentials. 3. The quantal content of endplate potentials was increased by BDM in a reversible, concentration-dependent manner. 4. The frequency of miniature endplate potentials was not changed by 10 mM BDM in the presence of normal or raised potassium concentrations, indicating that BDM does not change quantal content by a direct effect on calcium channels or on steady-state intracellular calcium concentration. 5. A change in the time course of the extracellularly recorded nerve terminal action potential caused by BDM was similar to the change produced by 4-aminopyridine (4-AP). 6. The increase in quantal content produced by BDM was only slightly reduced in the presence of 1 mM tetraethylammonium (TEA) but was significantly reduced in the presence of 0.5 to 1 mM 4-AP. 7. It was concluded that BDM blocks a 4-AP-sensitive potassium conductance in motor nerve terminals and, by increasing the duration of the action potential in this way, increases evoked transmitter release.

Effects of 2,3-butanedione monoxime on blood pressure, myocardial Ca2+ currents, and action potentials of rats.

2,3-Butanedione monoxime (BDM) has a negative inotropic effect on smooth muscles as well as the myocardium. Therefore, in the present study we compared the sensitivity to BDM of the cardiovascular system of the spontaneously hypertensive (SHR) and the normotensive Wistar-Kyoto (WKY) rat. While BDM significantly decreased the blood pressure (BP) for both strains, the SHR was significantly more responsive. Specifically, 5, 30, 100, and 200 mg/kg BDM (intravenously) reduced BP of the SHR by 9 +/- 3, 20 +/- 3, 49 +/- 5, and 63 +/- 7 mm Hg, respectively. The same doses of BDM reduced BP of the WKY by 0, 2 +/- 0.4, 18 +/- 3, and 26 +/- 3 mm Hg. The duration of the hypotensive effect of BDM was also greater for the SHR. In vitro, BDM had a greater suppressant effect on the L-type Ca2+ current (ICa(L)) of SHR ventricular myocytes; the IC50 for the suppression of ICa(L) was 17 and 29 mM for SHR and WKY ventricular myocytes, respectively. The beta-adrenergic receptor agonist isoproterenol antagonized the suppressant effect of BDM on ICa(L). Furthermore, BDM significantly reduced the duration of both spontaneous and electrically stimulated action potentials of cultured neonatal rat cardiomyocytes. Intracellular dialysis with the catalytic unit of protein kinase A antagonized BDM's effect on the action potential. These data suggest that suppression of myocardial activity contributes to the hypotensive effect of BDM. In addition, the elevated response to BDM of SHR cardiac myocytes may indicate that the conformation and/or modulation of the L-type Ca2+ channel differ for the SHR and WKY lines of rat.

Ergotism: a possible etiology for puerperal psychosis.

Some reports in the medical literature have mentioned the occurrence of psychotic reactions in response to the use of certain ergot alkaloids in therapeutic doses. Prompted by these observations, we undertook a search for cases of "pure" puerperal psychosis (ie, typical manifestations 3-14 days postpartum) in order to evaluate the clinical background of this phenomenon. Special attention was paid to the medications that the patients had received peripartum. In the last 10 years, out of eight perinatal centers, we found only three cases that fulfilled the criteria of the quoted entity. In all instances, the manifestations of puerperal psychosis had been preceded by the administration of ergot derivatives. Based on the presented data, we hypothesize that typical postpartum psychosis may represent an idiosyncratic reaction to potent vasoactive drugs including ergot derivatives. The similarities between the clinical manifestations of ergotism and puerperal psychosis, and some of the epidemiologic features of the latter condition, appear to implicate ergot alkaloids as potential causative agents. Although the validity of the suggested interpretation requires further evaluation, we believe that the currently available data warrant caution with regard to the administration of ergot derivatives postpartum. These drugs should not be used in the absence of clear indication or in unnecessarily high doses. We suggest that ergotism be included in the differential diagnosis in cases of pure puerperal psychosis.

Toxins selective for subunit interfaces as probes of nicotinic acetylcholine receptor structure.

The pentameric structure of the nicotinic acetylcholine receptor with two of the five subunit interfaces serving as ligand binding sites offers an opportunity to distinguish features on the surfaces of the subunits and their ligand specificity characteristics. This problem has been approached through the study of assembly of subunits and binding characteristics of selective peptide toxins. The receptor, with its circular order of homologous subunits (alpha gamma alpha delta beta), assembles in only one arrangement, and through mutagenesis, the residues governing assembly can be ascertained. Selectivity of certain toxins is sufficient to readily distinguish between sites at the alpha gamma and alpha delta interfaces. By interchanging residues on the gamma and delta subunits, and ascertaining how they interact with the alpha-subunit, determinants forming the binding sites can be delineated. The alpha-conotoxins, which contain two disulfide loops and 12-14 amino acids, show a 10,000-fold preference for the alpha delta over the alpha gamma subunit interface with alpha epsilon falling between the two. The waglerins, as 22-24 amino acid peptides with a single core disulfide loop, show a 2000-fold preference for alpha epsilon over the alpha gamma and alpha delta interfaces. Finally, the 6700 Da short alpha-neurotoxin from N. mossambica mossambica shows a 10,000-fold preference for the alpha gamma and alpha delta interfaces over alpha epsilon. Selective mutagenesis enables one to also distinguish alpha-neurotoxin binding at the alpha gamma and alpha delta subunits. This information, when coupled with homology modeling of domains and site-directed residue modification, reveals important elements of receptor structure and conformation.

Advantages of the triangularis sterni muscle of the mouse for investigations of synaptic phenomena.

The triangularis sterni muscle (TS) of the mouse is a thin trapezoidal sheet of fibres in which individual neuromuscular junctions are easily observed with Nomarski optics. Thus, microelectrodes are readily positioned to accurately record various synaptic phenomena. For example, miniature end-plate currents were easily recorded with a focally positioned extracellular electrode and the end-plate sensitivity to acetylcholine averaged 2062 mV/nC. In addition, the intercostal nerves segmentally innervate the TS. Electrophysiologic and histologic analysis showed that each nerve innervates a sharply defined territory of the muscle surface. These preliminary observations suggest that the TS may be ideal for studies of synaptic function and the processes underlying synapse stabilization in the mammal.

Prediction of major depression and dysthymia from CES-D scores among ethnic minority adolescents.

OBJECTIVE: The Native Hawaiian Mental Health Research Development Program is an epidemiological longitudinal study of adolescents residing in Hawaii. This article examines the utility of the Center for Epidemiologic Studies-Depression Scale (CES-D) for predicting DSM-III-R diagnoses of major depression (MD) and dysthymic disorder (DD) and investigates whether prediction differs by gender and ethnicity. METHOD: Diagnostic Interview Schedule for Children interviews were conducted with 556 adolescents randomly selected from among more than 7,000 students who had completed the CES-D. RESULTS: Six-month prevalence rates were as follows: MD = 8.5%, DD = 4.7%, either (MDDD) = 9.9%. Prevalence rates were significantly higher among females, but after CES-D scores were accounted for, gender no longer predicted depression in most analyses. When a cutoff score of 16 was used, classification accuracy was lower for Native Hawaiians than non-Hawaiians. However, after group differences in gender and grade level were accounted for, the predictive validity of the CES-D did not differ by ethnicity. CES-D factor 1 scores identified MD, DD, and MDDD about as well as the total score or all three factors together. CONCLUSIONS: These results support the validity of the CES-D for screening for depression among adolescents of Native Hawaiian and other minority backgrounds.

2,3-Butanedione monoxime modifies the glycine-gated chloride current of acutely isolated murine hypothalamic neurons.

In this study, we explored the effect of the chemical phosphatase 2,3-butanedione monoxime (BDM) on glycine current (IGly) of murine ventromedial hypothalamic neurons. Co-application of 0.01 to 67 mM BDM increased IGly decay rate with little change of the peak amplitude. This effect was both rapid in onset and offset and required the presence of the agonist. Pretreatment with BDM alone did not alter-IGly decay. In addition, dialysis of neurons with 500 microM ATP-gamma-S did not alter the acute effect of BDM. Thus, this effect may result from open channel block rather than BDM-induced dephosphorylation of the receptor/channel protein. In contrast to the acute effect described above, relatively prolonged (i.e., greater than 80 s) pretreatment with BDM reduced peak IGly. The phorbol ester (PDBu), a protein kinase C (PKC) activator, mimicked this effect of BDM. Furthermore, chelerythrine, a specific PKC inhibitor, prevented this effect of BDM on peak IGly. Thus, activation of PKC may mediate this attenuating effect of BDM on IGly. For a sub-population of these pretreated neurons, there was a subsequent potentiation of IGly which followed the initial suppressant effect. This potentiation may be due to a phosphatase effect of BDM, since it was observed more frequently when neurons were also pretreated with the protein kinase inhibitors H7 or chelerythrine. These findings suggest that BDM alters protein kinase activity and acts as a phosphatase to regulate the activity of the glycine receptor/channel complex.

Chronic ingestion of ethanol increases the number of Ca2+ channels of hippocampal neurons of long-sleep but not short-sleep mice.

Whole cell and single channel Ca2+ currents were compared for hippocampal neurons of normal and ethanol-tolerant long-sleep and short-sleep mice. The properties of these currents were equivalent for normal LS and SS mice. In contrast, the peak amplitude of the whole cell Ca2+ current increased significantly for neurons isolated from LS but not SS mice chronically ingesting ethanol. Since there were no changes in the functional properties of single channel events, these data indicate that chronic ingestion of ethanol causes an increase in the number of functional Ca2+ channels in the membrane of hippocampal neurons of LS mice.

Interaction of the opiate antagonist, naltrexone methyl bromide, with the acetylcholine receptor system of the motor end-plate.

Naltrexone methyl bromide (NMB) caused alterations in both the amplitude and the time course of end-plate currents (EPC) recorded from rat muscle. The amplitudes were significantly lower than control, and the EPC decay was no longer singly exponential. There was also a decrease in the voltage dependence of the peak amplitude and the decay time constant of the EPCs. These findings suggest NMB is blocking the open state of the acetylcholine channel.

cAMP-dependent protein kinase modulation of glycine-activated chloride current in neurons freshly isolated from rat ventral tegmental area.

Adenosine 3',5'cyclic monophosphate-(cAMP)-dependent protein kinase (PKA) modulation of glycine-activated Cl- currents (IGly) in single neurons freshly isolated from the rat ventral tegmental area (VTA) was studied using whole-cell patch-clamp technique. In the majority of cells tested with Mg-ATP in the internal solution, IGly induced by 3-10 microM glycine increased spontaneously (ran up). In the absence of internal ATP, IGly remained stable in six of seven cells. External perfusion of 8-Br-cAMP, a PKA activator, potentiated IGly only in cells showing run-up. 8-Br-cAMP potentiated IGly induced by low concentrations of glycine, but had no effect on the maximal current. When added to the pipette solution, H-89, a PKA inhibitor, blocked ATP and 8-Br-cAMP induced run-up of IGly. In contrast, dialysis with chelerythrine, a PKC inhibitor, did not alter the run-up of IGly. These results suggest that the PKA pathway modulates the activity of the glycine receptor/channel complex via enhancing the affinity of the receptor for glycine.

Waglerin-1 inhibits GABA(A) current of neurons in the nucleus accumbens of neonatal rats.

The effect of Waglerin-1, a 22-amino acid peptide purified from the venom of Wagler's pit viper on the whole cell current response (I(GABA)) to gamma-aminobutyric acid (GABA) was examined for neurons freshly isolated from the nucleus accumbens of 3- to 7-day-old rats. Waglerin-1 depressed I(GABA) induced by subsaturating concentrations of GABA; the IC(50) for I(GABA) induced by 10 microM GABA was 2.5 microM Waglerin-1. This concentration of Waglerin-1 shifted the GABA concentration-response curve to the right in a parallel manner, increasing the GABA EC(50) from 12+/-3 to 27+/-5 microM. The depressant effect of Waglerin-1 was greater at negative holding potentials. Zn(2+) also inhibited I(GABA) with an IC(50) of 0.3 microM. Phosphorylation state appeared to modulate GABA(A) receptor sensitivity to the inhibitory effect of Waglerin-1 since dialysis of neurons with N-[2-((p-bromocinnamyl)amino)ethyl]-5-isoquinolinesulfonamide HCl (H-89), an inhibitor of protein kinase A, prevented inhibition. The data are discussed in terms of developmental influences on the subunit composition of GABA(A) receptors in neurons of the nucleus accumbens.

Phosphorylation modulates the activity of the ATP-sensitive K+ channel in the ventromedial hypothalamic nucleus.

Regulation of the ATP-sensitive K+ (K-ATP) channel was examined in cell-attached and inside-out membrane patches of freshly isolated neurons from the ventromedial hypothalamic nucleus (VMN) of 7-14 day old male Sprague-Dawley rats. When inside-out patches were exposed to symmetrical K+, the reversal potential was -2.85 +/- 1.65 mV, the single channel conductance 46 pS, and the total conductance varied as a multiple of this value. Glucose (10 mM) reversibly inhibited channel activity in cell-attached preparations by 81%. In the presence of 0.1 mM ADP, 10, 5, and 1 mM ATP reversibly inhibited VMN K-ATP channels in inside-out patches by 88, 83, and 60%, respectively. This inhibition was not dependent on phosphorylation since 5 mM AMPPNP, the non-hydrolyzable analog of ATP, reversibly inhibited channel activity by 67%. Relatively high concentrations of glibenclamide (100 microM) also reversibly inhibited VMN K-ATP channel activity in cell attached and inside-out patches by 67 and 79%, respectively. Finally, the non-specific kinase inhibitor H7 (200 microM) decreased channel activity by 53% while the non-specific phosphatase inhibitor microcystin (250 nM) increased channel activity by 218%. These data suggest that while the inhibitory effect of ATP is not phosphorylation dependent, phosphorylation state is an important regulator of the VMN K-ATP channel.