|
|
||||||||
1 Division of Respiratory Medicine, Department of Pediatrics, Yale University School of Medicine, New Haven, Connecticut 06520; and 2 Department of Biochemistry and Molecular Biology and Physiology, Institute of Biology and Molecular Genetics, Universidad de Valladolid, Consejo Superior de Investigaciones Cientificas, 47005 Valladolid, Spain
| |
ABSTRACT |
|---|
|
|
|---|
A preparation was developed that allows for the recording of single-unit chemoreceptor activity from mouse carotid body in vitro. An anesthetized mouse was decapitated, and each carotid body was harvested, along with the sinus nerve, glossopharyngeal nerve, and petrosal ganglia. After exposure to collagenase/trypsin, the cleaned complex was transferred to a recording chamber where it was superfused with oxygenated saline. The ganglia was searched for evoked or spontaneous unit activity by using a glass suction electrode. Single-unit action potentials were 57 ± 10 (SE) (n = 16) standard deviations above the recording noise, and spontaneous spikes were generated as a random process. Decreasing superfusate PO2 to near 20 Torr caused an increase in spiking activity from 1.3 ± 0.4 to 14.1 ± 1.9 Hz (n = 16). The use of mice for chemoreceptor studies may be advantageous because targeted gene deletions are well developed in the mouse model and may be useful in addressing unresolved questions regarding the mechanism of chemotransduction.
carotid body chemoreceptor; knockout; hypoxia
| |
INTRODUCTION |
|---|
|
|
|---|
CAROTID BODY CHEMORECEPTORS transduce a decrease in arterial oxygen tension into an increase in spiking activity on some carotid sinus nerve afferent fibers. The mechanism of transduction has remained enigmatic but is presumably dependent on the release of an excitatory transmitter from the glomus cells, a secretory cell apposed to the afferent nerve ending. However, the mechanism of oxygen sensing and identification of the purported transmitters have been problematic, because, in general, applications of antagonists to purported transmitters or oxygen sensors have produced variable and inconsistent results (16). The use of mice with targeted gene disruptions may offer an alternative approach to pharmacological interventions and give better specificity. Indeed, these mice have been used in the study of ventilatory changes during hypoxia or hyperoxia (18). However, ventilatory measurements are indirect, and the use of mice to understand transduction better is dependent on our ability to access chemoreceptor function directly.
In 1982, Biscoe and Pallot (4) undertook the first recordings of mouse chemoreceptors in normal and wobbler-mutant mice using a wire recording of the sinus nerve in vivo. Because of the small size of the sinus nerve, these recordings were technically difficult, unstable, and generally had multiple modalities. Furthermore, the small size of the sinus nerve obviated an ability to further dissect the nerve to obtain better resolution of unit activity or modal specificity.
The present work was undertaken to develop a model by which high-resolution, single-unit chemoreceptor recordings may be undertaken from mouse chemoreceptors in vitro. Here we report a modification of a technique that our laboratory has previously employed to record rat chemoreceptors in vitro (8). For this, the carotid body is harvested along with the petrosal ganglia, and extracellular recordings are obtained by using electrodes placed in the petrosal ganglia. Initial experiments demonstrated good stability and exceptional signal-to-noise ratio (signal/noise) compared with other recording methods.
| |
METHODS |
|---|
|
|
|---|
Initial carotid body isolation. Mice (6-8w, wild type, CD-1; Charles River) were deeply anesthetized by placement in a closed chamber with an atmosphere that was saturated with methoxyfluorane vapor. After deep anesthesia, as evidenced by an absence of motor movements (outside of breathing) and an absence of withdrawal reflex, the mice were removed and decapitated. The carotid body and petrosal ganglia were harvested under a dissecting microscope with the use of the following approach. During the following procedure, the exposed area was washed with cold saline to prevent desiccation.
1) The carotid arteries were dissected free and laterally retracted, allowing exposure of the carotid bifurcation and nodose ganglia. 2) The superior cervical ganglia was gently pulled free with fine forceps (Dumont no. 5) and cut free of the bifurcation. 3) The bone caudal to the entry point of the hypoglossal nerve was cut with a sharp scalpel blade (no. 10), and the bone was gently pried apart to allow for better access to the nodose/petrosal ganglia. 4) The internal carotid artery was cut at the point of its secondary bifurcation. 5) The vagus nerve next to the cut common carotid artery was dissected free to the nodose ganglia. The pharyngeal branch of the vagus was cut at its junction with the main branch of the vagus. 6) By gently pulling on the vagus nerve, the glossopharyngeal nerve and petrosal ganglia were dissected free of the surrounding tissue. After cutting centrally to the petrosal ganglia, the ganglia and glossopharyngeal nerve were reflected over the carotid bifurcation. 7) By gently pulling on the common carotid artery, the carotid bifurcation and glossopharyngeal nerve with attached ganglia were pulled away from the surrounding tissue and cut free. The bifurcation was placed in cold, oxygenated (95% O2-5% CO2) saline (in mM: 120 NaCl, 3 KCl, 2 CaCl2, 1 Na2HPO4, 1 MgSO4, 24 NaHCO3, and 10 glucose). 8) By using sharp dissection, any muscle tissue was removed, and the common carotid and external carotid arteries were dissected free. The remaining portion of the bifurcation and glossopharyngeal nerve was placed in a 35-mm culture dish filled with dilute solutions of collagenase (0.01%, type P; Boeheringer) and protease (0.005%, type IX; Sigma Chemical) dissolved in oxygenated saline at room temperature for 30 min. The dish was gently rocked on a shaker table. 9) At the end of the enzyme exposure, the chemoreceptor complex was transferred to fresh oxygenated saline. The remaining connective tissue was gently trimmed from around the carotid body. In this case, it is often easier to identify the sinus nerve as the first nerve to exit from the glossopharyngeal nerve and clean the sinus nerve toward the carotid body. 10) The cleaning step was the most difficult because the sinus nerve is quite small and fragile (~20-µm diameter) and up to 1 mm in length (Fig. 1). Transfer of the chemoreceptor complex is by aspiration into a large-bored pipette with care taken to minimize pulling the carotid body with the sinus nerve.
|
Nerve recording and unit identification.
The chemoreceptor complex was transferred to a recording chamber
(volume
250 µl) placed on an inverted microscope and superfused at
2.5 ml/min with heated (32-33°C) saline equilibrated with 20 or 95% O2-5% CO2-balance N2. The
lower temperature (compared with body temperature) was selected because
viability may be prolonged at the lower temperature and transduction of
hypoxic stimuli still occurs (17). A metal microelectrode (10 M
;
Frederick Haer, Brunswick, ME) was advanced into the carotid body,
which held the carotid body in place as well as provided a means of
initiating an orthodromic action potential. A second microelectrode was
advanced into the petrosal ganglia and used to hold the ganglia in
place in the perfusion chamber.
Experimental protocol. Once a purported chemoreceptor unit was observed based on orthodromic electrical activation, the electrical stimulus was removed, and spontaneous unit activity was measured. The response to acute hypoxia was tested by switching the perfusate from normoxia to saline equilibrated with 5% CO2-balance N2 for 2 min. Without an oxygen scavenger, chamber PO2 fell over a period of ~1 min and reached a nadir of ~20 Torr.
Data analysis. The time period between an orthodromic electrical stimuli and arrival of the orthodromic spike was digitized (pCLAMP, Axon Instruments, Foster City, CA) and measured to give an estimate of the conduction velocity based on the latency between the stimulus artifact and arrival of the orthodromic action potential. Signal/noise was calculated based on the ratio of the peak voltage change from baseline during the action potential to the standard deviation of the resistive noise recorded in the absence of spikes. In some cases, samples of nerve activity during normoxia and during hypoxia were digitized at 6 kHz and recorded to disk (Axotape, Axon Instruments).
Spontaneous spiking activity was converted to logic pulses by a window discriminator (DIS-1, BAK Instruments) and summed each second (TECMAR Labmaster, Scientific Solutions). Baseline spontaneous spiking activity was measured over 100 s during superfusion with normoxic saline (PO2
150 Torr) and during the
peak discharge over 3 s during hypoxic stimulation
(PO2
20 Torr).
In some cases, the time period between neighboring spikes was timed to
1-ms accuracy and measured for ~1,000 consecutive spikes. A subgroup
of 500 consecutive intervals was selected on the basis of a
steady-state mean discharge frequency and used to construct an interval
histogram. Counts in the histogram bins were compared with those
expected for a random, Poisson-type process by using a
2
statistic, as previously described (5, 22).
| |
RESULTS |
|---|
|
|
|---|
Evoked activity and signal/noise.
Identification of purported chemoreceptor afferent cells was undertaken
by applying a brief electrical stimulus (100-500 µA, 0.5-ms
duration) through a stimulating electrode placed in the carotid body.
Evoked orthodromic activity was recorded by using a suction electrode
placed in the petrosal ganglia. An example of the evoked action
potential is shown in Fig. 2 in which the stimulus artifact is followed by an orthodromic spike.
|
Spiking activity during normoxia and hypoxia.
All 16 purported chemoreceptor afferents had spontaneous spiking
activity during normoxia (Fig. 3). On
average, spiking activity in normoxia was 1.3 ± 0.4 Hz. After the
perfusate was switched to saline equilibrated with 5%
CO2-balance N2, chamber
PO2 fell to near 20 Torr over the
course of 60-80 s (Fig. 3). Peak, single-fiber spiking activity in
hypoxia was 14.1 ± 1.9 Hz (n = 16). The relationship between
chamber PO2 and spiking activity
showed the typical exponential relationship (Fig.
4). Chamber
PO2 at which the nerve activity
increased to one-half of the peak level was 45 ± 4 Torr (n = 14).
|
|
Pattern of afferent spiking activity from mouse chemoreceptors.
In four cases, the interspike interval periods were timed to 1-ms
accuracy and used to calculate interspike-interval histograms and
serial correlation coefficients. In these four cases, a subsample of
500 consecutive intervals was selected based on a constant average
discharge frequency. In all four cases, the interspike-interval histogram was not significantly different from exponential (Fig. 5), suggesting that the pattern generator
was not significantly different from a Poisson-type random process.
|
| |
DISCUSSION |
|---|
|
|
|---|
The present results demonstrate the feasibility of recording single-unit chemoreceptor activity with high signal/noise from the mouse carotid body and provide preliminary observations showing that spike generation in mouse chemoreceptors is similar to that of other species. The "chemoreceptor complex" preparation (carotid body, sinus nerve, glossopharyngeal nerve, petrosal ganglia) offers significant advantages over conventional nerve recording with excellent stability and high signal/noise.
Experimental preparation. This experimental preparation is similar to that which our laboratory originally developed for rat ganglion recordings of chemoreceptor afferent fibers (8). This, in turn, was a variation of that originally developed by Belmonte and colleagues (3) for intracellular recordings of cat petrosal neurons with carotid body projections. In both cases, the carotid bodies were harvested along with the petrosal ganglia and maintained in vitro. In our model, we used glass suction electrodes to obtain extracellular unit activity from the soma of the petrosal neurons. This was based on the previous success of Vidruk and Dempsey (25) and Mulligan et al. (20), who used tungsten microelectrodes to obtain extracellular unit recordings from petrosal neurons of pig and cat in vivo.
Harvesting of the petrosal ganglia along with the carotid body offers several advantages. First and foremost, it allows single-unit discrimination that is far beyond that obtained by using nerve recordings and allows for quantifying chemoreceptor activity based at a single unit level. Multiunit activity, as is generally recorded from the whole or partial sinus nerve, gives no information regarding the magnitude of fiber recruitment, magnitude of contamination due to recording noise, or magnitude of nonchemoreceptor spiking. These problems may be especially difficult to avoid in the mouse, because the nerve is too small to allow for dissection into smaller strands (4). The improved signal/noise of the somal recording compared with the axon recording may be readily appreciated by comparing the oscillographic traces of Figs. 2, 3, and 5 to previous recordings by our laboratory (17) and others (24) from rat sinus nerve. The improvement in signal/noise is due to not only the greater electrical field around the soma of afferent neurons (because of their larger size) but also the physical separation from the field of other chemoreceptor afferents. Background spiking activity from neighboring somas was generally not observed in our recordings (Figs. 3 and 5).Mouse chemoreceptor discharge characteristics. All of the discharge characteristics from the mouse are consistent with those reported previously for the rat and are generally consistent with those obtained in larger species (cat, dog, pig, goat). Baseline spiking activity was ~1 Hz, and peak rates were ~14 Hz in the mouse. Similar baseline and peak rates were previously recorded in the rat under similar conditions, but the peak rate in the rat was slightly higher, perhaps because of the use of an oxygen scavenger in the previous rat studies (7, 17). The increase in spiking activity occurred primarily at <80-Torr PO2, compared with ~150 Torr in the rat (24). However, this is to be expected because the rat carotid body is larger (increasing the diffusion gradient) and the previous experiments were done at a higher temperature.
The spike generation property of mouse chemoreceptors is also consistent with that of other species. The interspike interval distribution was exponential, suggesting that the generation process was random, i.e., like the process of radioactive decay. A similar result was previously obtained from the analysis of afferent spike generation of rat chemoreceptors in vitro (5) and one single-fiber recording of mouse chemoreceptors in vivo (4). In other species, interspike intervals also approximate a random process having a mean frequency that is modulated by the respiratory and cardiac cycles (15). Of course, this modulation would be absent in vitro. In addition, some chemoreceptors, particularly in birds and goats, produce doublet spikes, which are manifest as a large number of counts in the first time bin of the interval histogram (21, 23). However, doublet bursting was not observed in our mouse recordings.Potential usefulness for enhanced understanding of the chemotransduction process. The mechanism by which the afferent spike is initiated at the afferent nerve terminal is presently unknown. Whereas it is widely assumed that an excitatory transmitter is involved, the identity and nature of the presumed chemical are unresolved. Whereas a complete assessment of the purported transmitters is beyond the present scope, the example of dopamine may illustrate the point. Dopamine is synthesized and stored by the glomus cell and is released by hypoxia in close temporal association with the increase in afferent nerve activity (6, 12). Furthermore, the messages for dopamine D1 and D2 receptors are expressed in both the carotid body and petrosal ganglia (1, 2, 14). This has been taken to argue that dopamine may serve an excitatory role, but other observations are seemingly inconsistent with this role. In particular, exogenous administration of dopamine generally inhibits ongoing nerve activity, and administration of dopamine antagonists increases afferent nerve activity (9, 19, 26). One may conclude either that catecholamine secretion is not causal to the increase in afferent nerve activity or that other dopamine receptors (e.g., autoreceptors) are the primary target for exogenous agonists and antagonists.
The use of genetically altered mice perhaps offers the optimal approach to resolve such issues. A pharmacological approach has an uncertainty of whether the agonist or antagonist reaches the desired site. In contrast, mice with specific targeted disruptions of dopamine D1, D2, D3, and D4 receptors have all been developed and can be maintained to adulthood (10). A direct assessment of chemoreceptor responsiveness in these mice might yield a unique insight into the role(s) of dopamine in chemoreceptor transduction. In a similar vein, the use of knockout animals has already been used to assess the role of neurokinin type 1 receptors (11) and neuronal nitric oxide synthase (18) in the ventilatory responses to hypoxia. A natural extension of these experiments would be an examination of the consequences of these knockouts on peripheral chemoreceptor function. In summary, the present report details a method of harvesting and obtaining single-unit chemoreceptor afferent recording in mouse carotid body in vitro. The results show spontaneous spiking rates during normoxia and hypoxia and a pattern of spike generation that is generally consistent with that obtained in rat carotid body in vitro and cat carotid body in vivo. It is hoped that the application of this model will ultimately prove useful for better understanding the mechanism of chemotransduction.| |
ACKNOWLEDGEMENTS |
|---|
The intellectual guidance and support of Constancio Gonzalez is gratefully acknowledged.
| |
FOOTNOTES |
|---|
This work was supported by National Heart, Lung, and Blood Institute Grant HL-46149.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Original submission in response to a special call for papers on "Hypoxia Influence on Gene Expression."
Address for reprint requests and other correspondence: D. F. Donnelly, Dept. of Pediatrics, Yale Univ. School of Medicine, 333 Cedar St., New Haven, CT 06520 (E-mail: David.Donnelly{at}Yale.edu).
Received 27 July 1999; accepted in final form 5 November 1999.
| |
REFERENCES |
|---|
|
|
|---|
1.
Bairam, A.,
Frenette J.,
Dauphin C.,
Carroll J. L.,
and
Khandjian E. W.
Expression of dopamine D1-receptor mRNA in the carotid body of adult rabbits, cats and rats.
Neurosci. Res.
31:
147-154,
1998[ISI][Medline].
2.
Bairam, A.,
and
Khandjian E. W.
Expression of dopamine D2 receptor mRNA isoforms in the carotid body of rat, cat and rabbit.
Brain Res.
760:
287-289,
1997[ISI][Medline].
3.
Belmonte, C.,
Gallego R.,
and
Morales A.
Membrane properties of primary sensory neurones of the cat after peripheral reinnervation.
J. Physiol. (Lond.)
405:
219-232,
1988
4.
Biscoe, T. J.,
and
Pallot D. J.
The carotid body chemoreceptor: an investigation in the mouse.
Q. J. Exp. Physiol.
67:
557-576,
1982
5.
Donnelly, D.
Generation of interspike intervals of rat carotid body chemoreceptors.
In: Frontiers in Arterial Chemoreception, edited by Zapata P.,
Eyzaguirre C.,
and Torrance R. W.. New York: Plenum, 1996, p. 169-174.
6.
Donnelly, D. F.
Electrochemical detection of catecholamine release from rat carotid body in vitro.
J. Appl. Physiol.
74:
2330-2337,
1993
7.
Donnelly, D. F.,
and
Doyle T. P.
Developmental changes in hypoxia-induced catecholamine release from rat carotid body, in vitro.
J. Physiol. (Lond.)
475:
267-275,
1994
8.
Donnelly, D. F.,
Panisello J. M.,
and
Boggs D.
Effect of sodium perturbations on rat chemoreceptor spike generation: implications for a Poisson model.
J. Physiol. (Lond.)
511:
301-311,
1998
9.
Donnelly, D. F.,
Smith E. J.,
and
Dutton R. E.
Neural response of carotid chemoreceptors following dopamine blockade.
J. Appl. Physiol.
50:
172-177,
1981
10.
Drago, J.,
Padungchaichot P.,
Accili D.,
and
Fuchs S.
Dopamine receptors and dopamine transporter in brain function and addictive behaviors: insights from targeted mouse mutants.
Dev. Neurosci.
20:
188-203,
1998[ISI][Medline].
11.
Dreshaj, I. A.,
Agani F. H.,
Krause J. E.,
Richer E.,
Martin R. J.,
and
Haxhiu M. A.
NK-1 receptors (R) are not required for changes in respiratory output induced by hypoxia or cyanide administration in mice (Abstract).
Am. J. Respir. Crit. Care Med.
159:
A45,
1999.
12.
Fidone, S.,
Gonzalez C.,
and
Yoshizaki K.
Effects of low oxygen on the release of dopamine from the rabbit carotid body in vitro.
J. Physiol. (Lond.)
333:
93-110,
1982
13.
Finley, J. C. W.,
Polak J.,
and
Katz D. M.
Transmitter diversity in carotid body afferent neurons: dopaminergic and peptidergic phenotypes.
Neuroscience
51:
973-987,
1992[ISI][Medline].
14.
Gauda, E. B.,
Shirahata M.,
and
Fitzgerald R. S.
D2-dopamine receptor mRNA in the carotid body and petrosal ganglia in the developing cat.
Adv. Exp. Med. Biol.
360:
317-319,
1994[Medline].
15.
Gehrich, J. L.,
and
Moore G. P.
Statistical analysis of cyclic variations in carotid body chemoreceptor activity.
J. Appl. Physiol.
35:
642-648,
1973
16.
Gonzalez, C.,
Almaraz L.,
Obeso A.,
and
Rigual R.
Oxygen and acid chemoreception in the carotid body chemoreceptors.
Trends Neurosci.
15:
146-153,
1992[ISI][Medline].
17.
Kholwadwala, D.,
and
Donnelly D. F.
Maturation of carotid chemoreceptor sensitivity to hypoxia: in vitro studies in the newborn rat.
J. Physiol. (Lond.)
453:
461-473,
1992
18.
Kline, D. D.,
Yang T.,
Huang P. L.,
and
Prabhakar N. R.
Altered respiratory responses to hypoxia in mutant mice deficient in neuronal nitric oxide synthase.
J. Physiol. (Lond.)
511:
273-287,
1998
19.
Lahiri, S.,
Nishino T.,
Mokashi A.,
and
Mulligan E.
Interaction of dopamine and haloperidol with O2 and CO2 chemoreception in carotid body.
J. Appl. Physiol.
49:
45-51,
1980
20.
Mulligan, E.,
Alsberge M.,
and
Bhide S.
Carotid chemoreceptor recording in the newborn piglet.
In: Arterial Chemoreception, edited by Eyzaguirre C.,
Fidone S. J.,
Fitzgerald R. S.,
Lahiri S.,
and McDonald D. M.. New York: Springer-Verlag, 1990, p. 285-289.
21.
Niu, W. Z.,
Engwall M. J. A.,
and
Bisgard G. E.
Two discharge patterns of carotid body chemoreceptors in the goat.
J. Appl. Physiol.
69:
734-739,
1990
22.
Nolan, W. F.,
Donnelly D. F.,
Smith E. J.,
and
Dutton R. E.
Nonrandom chemoreceptor activity during superfusion in vitro.
Brain Res.
292:
194-197,
1984[Medline].
23.
Nye, P. C. G.,
and
Powell F. L.
Steady state discharge and bursting of arterial chemoreceptors in the duck.
Respir. Physiol.
56:
369-384,
1984[Medline].
24.
Pepper, D. R.,
Landauer R. C.,
and
Kumar P.
Postnatal development of CO2-O2 interaction in the rat carotid body in vitro.
J. Physiol. (Lond.)
485:
531-541,
1995[ISI][Medline].
25.
Vidruk, E. H.,
and
Dempsey J. A.
Carotid body chemoreceptor activity as recorded from the petrosal ganglion in cats.
Brain Res.
181:
455-459,
1980[Medline].
26.
Zapata, P.
Effects of dopamine on carotid chemo- and baroreceptors in vitro.
J. Physiol. (Lond.)
244:
235-251,
1975
This article has been cited by other articles:
![]() |
D. F. Donnelly Spontaneous action potential generation due to persistent sodium channel currents in simulated carotid body afferent fibers J Appl Physiol, May 1, 2008; 104(5): 1394 - 1401. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Prieto-Lloret, D. F. Donnelly, A. J. Rico, R. Moratalla, C. Gonzalez, and R. J. Rigual Hypoxia transduction by carotid body chemoreceptors in mice lacking dopamine D2 receptors J Appl Physiol, October 1, 2007; 103(4): 1269 - 1275. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. F. Donnelly Orthodromic spike generation from electrical stimuli in the rat carotid body: implications for the afferent spike generation process J. Physiol., April 1, 2007; 580(1): 275 - 284. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. V. S. Faustino and D. F. Donnelly Lamotrigine and phenytoin, but not amiodarone, impair peripheral chemoreceptor responses to hypoxia J Appl Physiol, December 1, 2006; 101(6): 1633 - 1640. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. V. S. Faustino and D. F. Donnelly An important functional role of persistent Na+ current in carotid body hypoxia transduction J Appl Physiol, October 1, 2006; 101(4): 1076 - 1084. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. F. Donnelly, I. Kim, C. Carle, and J. L. Carroll Perinatal hyperoxia for 14 days increases nerve conduction time and the acute unitary response to hypoxia of rat carotid body chemoreceptors J Appl Physiol, July 1, 2005; 99(1): 114 - 119. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. D. Kline, M. C. F. Buniel, P. Glazebrook, Y.-J. Peng, A. Ramirez-Navarro, N. R. Prabhakar, and D. L. Kunze Kv1.1 Deletion Augments the Afferent Hypoxic Chemosensory Pathway and Respiration J. Neurosci., March 30, 2005; 25(13): 3389 - 3399. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Izumizaki, M. Pokorski, and I. Homma Role of the carotid bodies in chemosensory ventilatory responses in the anesthetized mouse J Appl Physiol, October 1, 2004; 97(4): 1401 - 1407. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |